University of Southern California Dissertations and Theses

An Experimental Study Of Effects Of Interest And Authority Upon Understanding Of Broadcast Information

(USC Thesis Other)

An Experimental Study Of Effects Of Interest And Authority Upon Understanding Of Broadcast Information

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content T h is d is s e rta tio n h as b een 61-6297
m ic ro film e d e x a c tly a s re c e iv e d
LIVINGSTON, H a ro ld M a u ric e , 1915-
AN E X PE R IM E N T A L STUDY O F E F F E C T S O F
IN TER EST AND AUTHORITY UPON UNDER
STANDING O F BROADCAST INFORM ATION.
U n iv e rsity of S outhern C a lifo rn ia , P h .D ., 1961
S p e e c h -T h e a te r
University Microfilms, Inc., Ann Arbor, Michigan
AN EXPERIMENTAL STUDY OF EFFECTS OF INTEREST
AND AUTHORITY UPON UNDERSTANDING OF
BROADCAST INFORMATION
by
Harold Maurice L ivingston
A D is s e r ta tio n P resented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In P a r t i a l F u lfillm e n t of the
Requirements fo r the Degree
DOCTOR OF PHILOSOPHY
(Speech)
June 1961
UNIVERSITY O F SO U TH ER N CALIFORNIA
GRADUATE SCHOO L
UNIVERSITY PARK
LOS ANGELES 7 . CALIFORNIA
This dissertation, written by
..............H arold. M SJH U .c_e_ Living s t on...............
under the direction of his Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by the Dean of
the Graduate School, in partial fulfillment of
requirements for the degree of
D O C T O R OF P H I L O S O P H Y
..................
^ Dean
Date. JuneJ .-!..9.6.1.....................................
DISSERTATION COMMITTEE
TABLE OF CONTENTS
Page
LIST OF T A B L E S ......................................................................................... iv
LIST OF FIGURES......................................................................................... vi
Chapter
I . THE PROBLEM AND LITERATURE................................................. 1
The Problem
Review of the L ite ra tu re
Summary
Preview of Remaining C hapters
I I . METHODS, T BCHNIQUES, AND PROCEDURES........................... 26
Methods
Techniques
Procedures
Summary
I I I . PRESENTATION AND INTERPRETATION OF DATA . . . . 64
Raw Data
T esting the F i r s t Null Hypothesis
T esting the Second Null Hypothesis
T esting the Third Null Hypothesis
T esting the Fourth Null Hypothesis
T esting the F if t h Null Hypothesis
T esting the S ix th Null Hypothesis
T esting the Seventh Null Hypothesis
Summary
IV. SUM M ARY, CONCLUSIONS, IMPLICATIONS, AND
RECOMMENDATIONS................................................................. 86
Summary
Primary Conclusions
Secondary Conclusions
i i
Page5
Im p licatio n s
Recommendations
BIBLIOGRAPHY....................................................................................................105
APPENDIXES.........................................................................................................110
i i i
LIST OF TABLES
T able Page
1. I n te r e s t Rating of the Two Recorded Talks . . . 30
2. Speaker I d e n tif ic a tio n and I n te r e s t Rating
of the Recorded T a l k s ................................................... 34
3. Speaker I d e n tif ic a tio n and I n te r e s t Rating
of the S elected T a l k ........................................................ 35
4. Scores on Test Based on P rio r Knowledge and
Chance .................................................. 40
5. Item D if fic u lty on Basis of P rio r Knowledge
and C h a n c e ............................................................................... 41
6. R e latio n sh ip of Matching by GPA to Scores
on R etention T e s t ............................................................ 45
7. Matched P airs of Male Subjects Used in This
E x p e rim e n t............................................................................... 52
8. Matched P airs of Female S ubjects Used in This
E x p e rim e n t............................................................................... 53
9. Studio Schedule fo r the 16 Groups of Subjects . 55
10. Paired t A nalysis of Scores on Test of
Understanding ...................................................................... 66
11. Chi-Square Test of I n te r e s t Levels of C ontrol
and Experimental Groups Who Heard the Radio
T a l k ............................................................................................ 69
12. Paired t A nalysis of Scores on Test of Under
standing of Those Matched P a irs of Subjects
Who Marked the Radio Talk as I n te re s tin g . . . 72
13. Paired t Analysis of Test Scores of Those
Matched P airs of S ubjects Where One Subject
Found the Talk Highly In te re s tin g and His
iv
T ab le Page
Matched C ounterpart Found the Talk I n te r e s tin g 75
14. Comparative I n t e r e s t Levels of Male and Female
S u b je c ts Used in t h i s E x p e rim e n t............................... 77
15. Comparative Levels of I n te r e s t in the Talk fo r
Freshmen and Sophomore S u b jec ts Used in This
E x p e r im e n t................................................................................ 79
16. Comparison of T est Scores of S u b jec ts Whose
ACE Scores Were in the S ix th Tenth or Above
with S u b jec ts Whose ACE Scores Were Below the
S ix th Tenth . . . ' ............................................................. 82
17. R e su lts of T estin g Seven Null Hypotheses . . . . 84
18. Table of Raw Data of Paired S u b j e c t s ............................... 126
19. Comparison of T est Sc ores and Level of I n te r e s t
R elated to S u b je c ts Seated in C h a irs, 2, 3,
and 5 as D esignated in Figure 1 . . . . . . . 128
20. Summary of R e su lts Made by S u b jec ts Grouped
According to Week of P a r t i c i p a t i o n ..............................129
v
5
LIST OF FIGURES
Figure Page
1. S tu d io se atin g arrangement fo r s u b je c ts and
a s s i s t a n t w ith p ro v isio n fo r reco rd in g
names of s u b je c ts by a s s i s t a n t ................................. 57
v i
CHAPTER I
THE PROBLEM AND LITERATURE
Most of our knowledge regarding the e f f e c t s of the
mass media of rad io and t e l e v i s i o n has been based on con
c lu s io n s drawn from nonexperim ental s tu d ie s . C a rl Hovland
c a lle d a t t e n t i o n to the lack of knowledge of the e f f e c t s of
mass communication:
Who says something i s u s u a lly as im portant as what
i s said in the d e term in atio n of the impact of the com
m unication. S u rp ris in g , however, the in flu en ce of t h i s
f a c t o r on the e f f e c tiv e n e s s of the mass media has not
been e x te n s iv e ly in v e s tig a te d . There are a large num
b er of s tu d ie s of communicators, but the emphasis in
th e se an aly ses i s u s u a lly on the c h a r a c t e r i s t i c s of the
communicators and t h e i r r o le s in s o c ie ty r a t h e r th an on
t h e i r e f f e c t s . i
In commenting upon the problems of measuring e f
f e c t s of mass media, Head lamented the f a c t t h a t "Research
on e f f e c t s i s beset w ith so many d i f f i c u l t i e s th a t few
sc h o la rly s tu d ie s have been c a r r ie d out on t h i s most impor-
o
ta n t asp ect of mass communications." Merton expressed
^Carl I. Hovland, " E ffe c ts of the Mass Media of
Communication," Handbook of S o c ia l Psychology, Gardner
Lindsey, e d i t o r (Cambridge: Adddson-Wesley P ublishing Com
pany, I n c ., 1954), I I , 1071.
2
Sydney W . Head, Broadcasting in America (Boston:
Houghton M if f lin Company, 1956)V" pT 4$Y.
p o in te d ly both the weaknesses of the ty p i c a l study of per
suasion through mass media and the problem of e s ta b lis h in g
v a lid in d ic e s of the e f f e c t s of propaganda or p ersu a sio n
upon the ra d io l i s t e n e r or the t e l e v i s i o n view er.
The Problem
Purpose of the study
D espite the ra p id development of t e l e v i s i o n in the
l a s t te n years and the subsequent d ec lin e of rad io network
p r e s t i g e , the number of ra d io s t a tio n s in the U nited S ta te s
has in crease d s t e a d i l y . 4 T his growth has r e s u lte d in more
emphasis on lo c a l programing w ith g r e a te r o p p o r tu n itie s f o r
the scheduling of in fo rm atio n al programs a t fav o rab le hours
on many ra d io s t a t i o n s . I t i s widely assumed to be in the
p ublic i n t e r e s t to dissem inate inform ation t h a t i s w ell-
p rep ared , can s u s ta in the i n t e r e s t of the l i s t e n e r , and i s
both understood and accepted by the ra d io audience. The
ra d io t a l k i s one type of program th a t i s commonly used f o r
the p r e s e n ta tio n of inform ation.
For many years th e re have been two major p o in ts of
view on which r a d io s t a t i o n managers have scheduled educa
t i o n a l t a l k s . One group of b ro a d c a ste rs has in s i s te d th a t
^Robert K. Merton, Mass P ersuasion (New York: Har
p er and B ro th e rs, 1946), pp. 4-7.
A f
^"Summary of Commercial B ro a d c a stin q ." B roadcasting.
LVII (September 14, 1959), 133.
the v o catio n of the speaker i s r e l a t i v e l y unimportant if
the inform ation i s we 11-s c rip te d and has high in te re s t- v a lu e
fo r the intended audience. T his group of s ta tio n managers
has suggested th a t the in f o rm a tio n -s p e c ia lis t should prepare
the broadcast m a te ria l but th a t i t s d e liv e ry on the a i r
should be delegated to a p ro fe s s io n a l s t a f f announcer.
Another group of s ta tio n managers has f e l t th a t the voice
of the in fo r m a tio n -s p e c ia lis t adds p re s tig e and a u t h o r it a
tiv e n e s s to the ra d io p re s e n ta tio n and t h a t th e re i s g re a te r
a tte n tiv e n e s s to the inform ation when the s p e c i a l i s t him self
is the speaker. These b ro a d c a ste rs p re fe r not to use the
voices of p ro fe s s io n a l s ta f f announcers f o r sp e c ia l informa
ti o n a l programs.
While some em p irical evidence may be drawn from stud
ie s of public address to support each p o in t of view, no ex
perim ental ra d io study has been re p o rte d th a t has attem pted
to ev a lu ate the comparative e f f e c tiv e n e s s of the p ro fe ssio n
a l b ro a d c a ste r with th a t of the su b je c t-m a tte r s p e c i a l i s t in
the d issem in a tio n of inform ation through a ra d io t a l k .
Communication re s e a rc h e rs have m aintained th a t the
mass media them selves are invested with an aura of p re s tig e
by a large p o rtio n of t h e i r audience, and th a t the media
confer s ta tu s on public is s u e s , s o c ia l movements, p erso n s,
and o rg a n iz a tio n s . Klapper has summarized the re se a rc h by
L a z a rsfe ld , Merton, Waples, B erelson, Bradshaw, and o th e rs
in d ic a tin g the public t r u s t and persuasive influence of mass
5
media in t h i s co u n try .
R eports on the p r e s tig e of the media and t h e i r pro
f e s s io n a l r e p r e s e n ta tiv e s have been of stu d ie s th a t were
nonexperim ental in method. In commenting on the aura of
p r e s tig e a t t r i b u t e d to ra d io , Klapper has made t h i s s t a t e
ment:
Radio has been a lle g e d to possess unique persuasive
c a p a b i l i t i e s because of i t s o fte n being the f i r s t
medium t o communicate given m a te ria l t o the audience,
and because of a group f e e lin g a lle g e d to be e x p e ri
enced by the audience. These l a s t two a l le g a tio n s are
n e ith e r supported nor disproved by e x is tin g em p iric a l
e v id en ce .°
I t was the purpose of t h i s study t o provide informa
ti o n w ith which b ro a d c a s te rs and e d u c a to rs could improve
t h e i r uses of ra d io t a l k s a s means of im parting inform ation
to the p u b lic . From the r e s u l t s of t h i s study the broad-
N
c a s t e r or ed u cato r should be able t o e s t a b l i s h advanced
c r i t e r i a f o r s e le c tin g speakers to p re se n t inform ative t a l k s
by ra d io . The c r i t e r i a may be based upon more than the in
t u i t i o n and e m p iric a l evidence t h a t p erfo rce inform them
now; they could be based upon lo g ic a l choice th a t has been
discovered through c o n tro lle d experim entation. B ro ad c asters
^Joseph T. K lapper, "Mass Media and P ersu asio n ,"
The P rocess and E ffe c ts of Mass Communication. Wilbur
Schramm, e d i t o r lUrbana; U n iv e rs ity of I l l i n o i s P ress, 1954),
pp. 313-314.
^Joseph T. K lapper, "The Comparative E ffe c ts of the
Various Media," The P rocess and E ffe c ts of Mass Communica
t i o n . Wilbur Sc hra mm, e d i t o r (Urbana: UnTve r s I t y of I l l i n -
o is P re ss, 1954), p. 105.
and e d u c a to rs should be able t o stre n g th e n t h e i r c r i t e r i a of
s e le c tio n w ith e x p e rim e n ta lly d eriv ed knowledge of the r e l a
t i v e e x te n ts t o which the to p ic of an inform ative ra d io t a l k
i s more or le s s im portant th a n the p r e s e n te r of the t a l k in
promoting i n t e r e s t and understanding among th e audience.
Statem ent of the problem
The problem of t h i s study was t o measure the e f
f e c t s of i n t e r e s t in the to p ic and the eth o s of the speaker
upon un d erstan d in g of an inform ative r a d io t a l k .
To t h i s end, the fo llo w in g n u ll hypotheses were
te ste d :
1. As measured by a r e te n tio n t e s t , understanding
of a r a d io t a l k on sleep f o r the c o n tr o l group
of s u b je c ts who heard a p h y sic ia n did not d i f f e r
s i g n i f i c a n t l y from t h a t of those s u b je c ts who
lis te n e d t o a new scaster.
2. The le v e l of i n t e r e s t re p o rte d by those sub- .
j e c t s who lis te n e d t o the t a l k d e liv e re d by the
s u b je c t- m a tte r s p e c i a l i s t did not d i f f e r s i g n i f
ic a n tl y from the le v e l of i n t e r e s t in the in
fo rm a tio n re p o rte d by the s u b je c ts who heard the
newsman.
3. Among matched p a i r s of s u b je c ts who marked the
t a l k as " in t e r e s ti n g " r a t h e r th an "highly i n t e r
e s tin g " understanding of the ra d io t a l k on sleep
fo r the c o n tro l group of su b je cts who heard the
s p e c ia lis t did not d i f f e r s ig n if ic a n tly from
th a t of the su b je cts who heard the new scaster.
The average scores on the re te n tio n t e s t f o r the
su b jects who marked the t a l k as "highly i n t e r
e s tin g " did not d i f f e r s ig n if ic a n tly from the
scores made by t h e i r matched co u n te rp a rts who
marked the t a l k as " in te r e s tin g ."
The le v e l of in t e r e s t in the inform ation re
ported by the group of male su b je c ts did not
d i f f e r s ig n if ic a n tly from the average le v e l of
i n t e r e s t rep o rted by the group of female sub
j e c t s .
The average le v e l of i n t e r e s t in the ra d io t a l k
in d ic ated by the freshman su b je c ts did not d i f
f e r s ig n i f ic a n t ly from th a t of the sophomore
su b je cts.
Among su b je cts who were matched by ACE placement
scores w ith in the same te n th of the d i s t r i b u t i o n
of scores on the ACE t e s t of sc h o la stic a p titu d e
in Oregon S ta te C ollege, the average scores on
the t e s t of understanding by those whose ACE
scores were below the six th te n th did not d i f f e r
s ig n if ic a n tly from those whose ACE scores were
in or above the six th te n th .
7
D e fin itio n s
Measure. Two measuring instrum ents were used in
t h i s study to compare matched groups of college stu d en ts who
were exposed to the ra d io t a l k as d e liv ered by the physician
and the new scaster. A t e s t of re te n tio n of the inform ation
was given composed of m ultiple-choice items and a th re e -
point r a tin g scale of the le v e l of i n t e r e s t also was admin
i s te r e d . Both ev a lu atio n s were rep o rted by the su b jects
immediately follow ing the p re s e n ta tio n of the recorded t a l k .
I n t e r e s t . This term as used in the experiment was
defined as the re a c tio n to the t a l k by each subject as re
f le c te d in h is choice recorded on a ra tin g scale ranging
from no i n t e r e s t to high i n t e r e s t in the co n ten t.
Topic. The topic chosen fo r t h i s experiment was in
form ation on sleep, with emphasis on sp e c ific f a c tu a l data
not regarded as common knowledge.
E thos. This term was borrowed from A ris to tle to
s ig n ify the l i s t e n e r 's re a c tio n to the c h a ra c te r of the
speaker as r e f le c te d in the two in tro d u ctio n s to the rad io
t a l k . Dr. Maloney rep resen ted the a u th o rity on the topic
presented; Chet Parker symbolized the ra d io r e p o rte r.
7 Lane Cooper, The Rhetoric pf A ris to tle (New York:
App l e t on-Ce ntury-Crof tV / Tnc'., 1 ^ 2 )', pp . 8-V.
8
U nderstanding. In the study, understanding was de
fin ed as the amount of inform ation re ta in e d by the subject
as recorded by h is score on the t e s t of re te n tio n of the in
form ation.
Informative ra d io t a l k . For the purpose of t h i s in
v e s tig a tio n , an inform ative rad io t a l k was defined as a ta lk
based upon a to p ic of general i n t e r e s t and con tain in g in fo r
mation on the to p ic th a t was not considered common know
ledge .
L im itatio n s of the study
The lo c a tio n of the in v e s tig a to r and the e x p e ri
mental design of t h i s study demanded th a t the su b je c ts be
conveniently a v a ila b le to the in v e s tig a to r and be c lo s e ly
matched f o r the purpose of v a lid comparison. The su b je cts
were drawn from the group of freshmen and sophomore students
of Oregon S ta te College who were en ro lle d in the basic
speech courses. By in v e s tig a tin g the college stu d e n ts who
had tak en standard t e s t s , i t was po ssib le to match the sub
j e c t s f o r both ACE placement score® and GPA in the previous
Q
term. Although a random sample of the gen eral population
®The ACE t e i s the college ap titu d e examination of
the American Council on Education. At Oregon S tate College
scores on the t e s t are recorded by te n th s on a scale of 1
through 10. The lowest te n th of the scores made on the ex
am ination are recorded as 1 and the h ig h est te n th of the
scores are recorded as 10.
% h e te r m GPA r e p r e s e n t s th e g r a d e - p o i n t a v e ra g e of
would have been more d e s ira b le th an lim itin g the study to
co lle g e stu d e n ts , the design of the study made t h i s imprac
t i c a l . In s e le c tin g the co lleg e su b je c ts of the experim ent,
care was ta k e n to include as broad a range of in te llig e n c e
as p o s s ib le .
Because the experim ental design was dependent on the
development of a r a d io t a l k th a t would be of i n t e r e s t to a l l
s u b je c ts , the length of the t a l k was lim ite d to e ig h t min
u te s . The p re v ity of the t a l k lim ite d the development of a
r e te n tio n t e s t th a t would have s u f f i c i e n t r e l i a b i l i t y and
v a l i d i t y to measure adequately the amount of understanding
of the ta lk by the s u b je c ts . P relim inary t e s t i n g was accom
p lish e d to s a t i s f y t h i s in v e s tig a to r th a t the measuring in
strument would measure the understanding with s u f f i c ie n t
d is c r im ir e tio n fo r the purpose of the in v e s tig a tio n .
Assumptions
The in v e s tig a to r attem pted to c re a te a s e ttin g fo r
the ra d io l i s t e n e r th a t would c lo s e ly approximate the normal
s it u a ti o n in which r a d io lis te n i n g is done. The present
study elim in a te d the classroom as a s e ttin g fo r the e x p e ri
ment in favor of a small com fortable ra d io stu d io . The sub
j e c t s lis te n e d to the ra d io program in small groups of fiv e
stu d en ts or le s s , w ith members of both sexes p rese n t in each
the student on the b a s is of a grade of A re p re se n tin g four
p o in ts , B re p re s e n tin g th re e p o in ts , C re p re s e n tin g two
p o in ts , and D re p re s e n tin g one p o in t.
10
grouping. The p relim in ary in s tr u c tio n s given the su b je c ts
were designed t o promote a rela x ed noncom petitive atmosphere
in the stu d io w ith no in d ic a tio n of the t e s t i n g procedures
t o follow involving the co n ten t of the ra d io program. A ll
su b je c ts in d ic a te d on a th r e e -p o in t scale th a t the rad io
program was e i t h e r " in te r e s tin g " or "highly i n t e r e s t i n g ."
The responses of the stu d e n ts s a t i s f i e d t h i s in v e s tig a to r
t h a t the s u b je c ts expressed normal r e a c tio n s to the recorded
broadcast and probably would have lis te n e d to the program i f
th ey had tuned to a s ta t i o n carry in g the ra d io t a l k of t h i s
study.
S ig n ifican ce of the study
A number of s tu d ie s have been made of the e f f e c t of
the speaker of a t a l k upon the acceptance of the t a l k by an
audience. D ie tric h in v e s tig a te d the r e l a t i v e e f f e c tiv e n e s s
of the dynamic and c o n v e rs a tio n a l modes of ra d io speaking in
in flu en c in g a t t i t u d e s . ^ Haiman stu d ied the e f f e c t of source
c r e d i b i l i t y upon s h i f t - o f - o p i n i o n . ^ In G reat B r ita in ,
Trenaman and Vernon rep o rted the r e s u l t s of se v e ra l stu d ie s
of r e la tio n s h i p s between understanding of and i n t e r e s t in the
^ J o h n E. D ie tric h , "The R elativ e E ffe c tiv e n e s s of
Two Modes of Radio D elivery in In flu e n cin g A ttitu d e s ,"
Speech Monographs. X II I s i (1946), 58-65.
■^Franklyn S. Haiman, "An Experim ental Study of the
E ffe c ts of Ethos in Public Speaking, Speech Monographs.
XVI (September, 1949), 191-92.
IX
12
programs of the B r itis h B roadcasting C o rporation. Kelman
and Hovland stu d ied s h if t- o f - o p in io n w ith th re e ty p e s of
a u th o rity to which groups of high school stu d e n ts were ex-
13
posed. Highlander re p o rte d an in v e s tig a tio n of speaker
v a r ia b le s in the p r e s e n ta tio n of ra d io t a l k s to l i s t e n e r s
whose responses were measured by an experim ental audience
an aly zer and by t r u e - f a l s e r e te n tio n t e s t s . ^
No study has been re p o rte d in which attem pt was made
to measure th e comparative e f f e c t s of i n t e r e s t in the to p ic
and the ethos of the speaker upon understanding of an in
form ative ra d io t a l k . Although the ra d io in fo rm atio n al t a l k
i s not popular among s ta tio n program m a n a g e rs ,^ i t i s the
the le a s t expensive form of p re se n tin g inform ation t o a
ra d io audience; because of i t s low c o s t i t i s probable th a t
i t w ill remain a s ig n i f ic a n t p a rt of the inform ative pro
graming of a rad io s ta tio n .
T his study should provide inform ation with which the
ra d io s t a t i o n manager could improve h i s use of ra d io t a l k s
12
Joseph Trenaman, "Understanding Radio T a lk s ," The
Q u a rte rly Jo u rn a l of Speech. XXXVII (A p ril, 1951), 173-7$.
■^Herbert C. Kelman and C a rl I . Hovland, "'R e in
sta te m e n t1 of the Communicator in Delayed Measurement of
Opinion Change," The Jo u rn a l of Abnormal and S o cial Psychol
ogy. XLVIII (Ju ly , 1953), 327-3b.
^ J o h n P. H ighlander, "Audience Analyzer Measure
ments and In fo rm atio n al E ff e c ts of Speaker V a ria b le s in
Radio T a lk s ," Speech Monographs. XXI (August, 1954), 188-89.
■^Giraud C h ester and Garnet R. G arrison, T e le v is io n
and Radio (2nd e d .; New York: A ppleton-C entury-C rofts, I n c . ,
12
f o r im parting inform ation to the p u b lic . From the r e s u l t s
of t h i s study, the b ro a d c a ste r or educator should be able to
e s t a b l i s h lo g ic a l c r i t e r i a fo r s e le c tin g speakers to p re se n t
inform ative t a l k s by r a d io . The p rese n t s e le c tio n of such
speakers i s based upon i n t u i t i o n and the e m p irical e x p e ri
ence of the s ta tio n manager. The s e le c tio n could be based
upon lo g ic a l choice th a t has been discovered through con
t r o l l e d exp erim en tatio n . T his study should enable the
b ro a d c a ste r or ed u cato r t o stren g th en h is c r i t e r i a of se le c
t i o n of both to p ic f o r the ra d io t a l k and the person desig
nated to d e liv e r the t a l k by the experim entally derived
knowledge of the com parative importance of the to p ic of a
t a l k and the eth o s of the speaker in promoting i n t e r e s t and
understanding among the audience.
Review of the L ite ra tu re
Statem ent of sources searched
In 1940 Ewbank re p o rte d t h a t t h i r t e e n of the 1186
th e s e s t h a t had been l i s t e d by Knower in Speech Monographs
I
were s tu d ie s of ra d io . ° F ifte e n years l a t e r W illis made
another survey of ra d io re se arc h l i s t i n g s in Speech Mono
graphs and the Dow surveys. W illis found seven d o c to ra l
1966), 289.
^ H e n ry L. Ewbank, "Trends in Research in Radio
Speech," The Q u a rte rly Jo u rn a l of Speech. XXVI (A pril,
1940), 286.
experim ental s tu d ie s t h a t were designed t o in v e s tig a te e le -
17
ments of ra d io communication. The Jo u rn a l of B roadcastinq
r e c e n tly p ublished a l i s t of 1,27b th e s e s and d i s s e r t a t i o n s
on ra d io and t e l e v i s i o n communication. Most of th e se stud
ie s were done under the su p erv isio n of u n iv e r s ity d ep art-
18
ments o th e r th a n Speech. D espite t h i s s u b s ta n tia l number
of in v e s tig a tio n s , few l i s t i n g s were of re se a rc h involving
the experim ental method.
In a d d itio n to th e se major sources, t h i s in v e s tig a
t o r examined P sychological A b s tr a c ts . ^ the jo u r n a ls of the
20
N ational A sso ciatio n of E ducational B roadcasters and the
N ational S o c ie ty f o r the Study of Communication, and the
p u b lic a tio n s c a ta lo g of the B r i t i s h B roadcasting Corpora-
22
tio n .
L ite ra tu re on source c r e d i b i l i t y
and i n t e r e s t f a c t o r s r e l a t e d to
speaker e f f e c tiv e n e s s
Hovland compared the r e l a t i v e e f f e c tiv e n e s s of the
^■^Edgar E. W illis , "Research in Radio and T e le v is io n
by Graduate S tu d en ts in Speech," The Q u a rte rly Jo u rn a l of
Speech. XLI (October, 195b), 266.
18
"Theses and D is s e rta tio n s in B ro ad castin g ," Jo u r
nal of B ro ad c astin g . I I (Winter, 1957-58), 57-90.
^ P s y c h o lo g ic a l A b s tr a c ts . XIV-XXXII (1940-58).
2qNAEB J o u r n a l. X-XVII (1950-58).
2^The Jo u rn a l of Communication. I-V III (1951-58).
00
B r i t i s h B roadcasting (London: B r i t i s h Broadcasting
C o rp o ratio n , 1954), pp. ■1-35.
ra d io commentary and the dramatic documentary program in
influencing opinion. The commentator was not id e n tif ie d in
the experim ent. Platoons of army tr a in e e s were studied as
su b je cts in comparing the p ersuasiveness of one-'sided versus
tw o-sided arguments on the p o ssib le length of World War I I .
Recorded p re s e n ta tio n s were played through a public address
system in to large auditorium s during army o rie n ta tio n p e r i
ods. T e sts were of both s h if t-o f-o p in io n and r e te n tio n of
inform ation. There were no s t a t i s t i c a l l y s ig n if ic a n t d i f
feren ce s in the e ffe c tiv e n e s s of the rad io commentary versus
23
the documentary program in t h i s study.
D ie tric h studied the r e la tiv e e ffe c tiv e n e s s of the
dynamic and c o n v e rsatio n a l modes of rad io speaking in in f lu
encing a t t i t u d e s . He used a speech designed to build
g re a te r lik in g fo r the Russian people and more confidence in
Russian i n t e g r i t y . Although six speakers recorded the
speech, the a u th o rity of each was not studied through the
experim ental design. D ie tric h rep o rted th a t the radio
speech did not r e s u l t in s ig n if ic a n t s h if t-o f-o p in io n . From
a p o s t- te s t of in te r e s t in the speech, he concluded th a t
w ithin the lim ita tio n s of h is experiment the amount of in
t e r e s t expressed in the program was " s ig n if ic a n tly r e la te d
2^Carl I. Hovland, A rthur Lumsdaine, and Fred Shef
f i e l d , Experiments on Mass Communication , Vol. I l l : S tudies
in S ocial Psychology in Worlcl War II (4 v o ls .; P rinceton:
P rinceton O niversixy t r e s s , pp. 230-35.
15
t o the e f f e c t of the program in in flu en c in g the s u b je c t's
a t t i t u d e s . " 24 This conclusion has some bearing on the p re s
ent in v e s tig a tio n .
Haiman studied the e f f e c t of source c r e d i b i l i t y upon
s h if t- o f - o p in io n on the su b ject of n a tio n a l compulsory
h e a lth in su ran ce, using w ire-recorded persuasive speeches
played before c la s s e s of stu d e n ts a t N orthwestern Univer
s i t y . The th re e speakers to whom the t a l k was a t t r i b u t e d
in the in tro d u c tio n s included the Surgeon-General of the
United S ta te s , the S ecretary-G eneral of the Communist Party
25
of America, and an anonymous u n iv e r s ity sophomore. Haiman
rep o rted t h a t the p re s tig e of the speaker in h is study did
influence the p ersu a siv en e ss of the speaker as shown by use
of the Woodward s h if t- o f- o p in io n b a l l o t . ^
Trenaman rep o rted the r e s u l t s of a number of B r itis h
s tu d ie s of r e la tio n s h i p s between understanding of programs
of the B r i t i s h B roadcasting C orporation and i n t e r e s t in the
programs. In one in v e s tig a tio n Trenaman used an inform ative
t a l k by a s c i e n t i s t on the su b ject of e le c tro n s . The t a l k
was f i f t e e n minutes in length and was follow ed by a r e te n
t io n t e s t of t h i r t y - t h r e e weighted item s. In co o p e ra tio n
with Trenaman, Vernon used inform ative t a l k s f o r an addi-
24D ie tric h , op. c i t . . p. 64.
25
Haiman, l o c . c i t .
26 I b i d . . p. 193.
16;
t i o n a l s e r ie s of s tu d ie s of i n t e r e s t and understanding. Al
though no p r e te s tin g of th e t a l k s was made in the s e r ie s of
experim ents, the su b je c ts were asked to in d ic a te i n t e r e s t in
the general to p ic p r io r to hearing the recorded t a l k or
a f t e r the t a l k had been p re se n te d . The major conclusion
th a t i s re le v a n t to the p re se n t study was on r e la tio n s h i p
between r e c a l l scores and i n t e r e s t r a t i n g s . Trenaman and
Vernon rep o rted th a t those who knew le s s about the to p ic s
of the t a l k s in d ic a te d more i n t e r e s t in the m a te ria l th a n '
those who were p a r t i a l l y informed on the to p ic s p re se n te d .
D ifferen ces in ed u c a tio n a l background were not the only de
term in an ts of the l i s t e n e r s ' understanding of ra d io informa
tio n although the persons of high in te llig e n c e tended to
make the b e s t scores on r e c a l l of the inform ation to which
a l l were e x p o s e d .^
Kelman and Hovland studied a p o s itiv e a u th o r ity , a
negative one and a n e u tra l one who spoke on the su b je ct of
ju v e n ile delinquency to c la s s e s of high school stu d e n ts.
The speaker was introduced on a t r a n s c r i p t i o n as a judge, as
a former ju v e n ile d e lin q u e n t, or as a member of the ra d io
audience. D iffe re n ces in voice and sty le of d e liv e ry were
introduced in keeping w ith the p e r s o n a l i t i e s suggested by
28
the in tro d u c tio n s . The co n ten t was not held c o n s ta n t fo r
27
Trenaman, loc. c i t .
^Kelm an and Hovland, op. c i t . , p. 330.
17
a l l th r e e p r e s e n ta tio n s , and the in v e s tig a to r s were not
s a t i s f i e d w ith the d e s ig n a tio n of the "n e u tra l" speaker in
the t a b u la t io n of s h if t- o f - o p in io n b a l l o t s . The high school
stu d e n ts tended to fav o r the speech given by the judge on
immediate a t t i t u d i n a l t e s t s , but the r e s u l t s were not s t a t
i s t i c a l l y s i g n i f i c a n t . 29
H ighlander re p o rte d an in v e s tig a tio n of speaker var
ia b le s in the p re s e n ta tio n of ra d io t a l k s to l i s t e n e r s
whose resp o n ses were measured by an experim ental audience
an a ly ze r and by a t r u e - f a l s e r e te n tio n t e s t . D iffe re n t in
tr o d u c tio n s of a speaker were the a u t h o r it y - p r e s t ig e v a r i
a b le s . C la s s e s of stu d e n ts who were e n ro lle d in fundamen
t a l s of speech courses lis te n e d to th e th irty -m in u te recorded
t a l k s on highway s a fe ty and p o lio m y e litis . O ther l i s t e n e r s
were drawn from p r o fe s s io n a l and s o c ia l groups in Madison,
W isconsin. W ithin the l i m i ta tio n s of the study, the le s s -
able speakers produced inform ation e f f e c t s equal t o those
produced by the more-able speakers. V a ria tio n in speaker
a u th o r ity - p r e s tig e d id not appear to be im portant in e s ta b
lis h in g the responses of the l i s t e n e r s . The fin d in g s sug
g e s t the conclusion t h a t the mass media tend to c o n fe r p re s
tig e and a u th o r ity on an in d iv id u a l by the very f a c t t h a t he
has been sin g led out to appear before the p u b lic . The audi
ence an aly ze r appeared to d i s t r a c t from the e d u c a tio n a l
29I b i d . . pp. 332-35.
fu n c tio n of the program. The le n g th s of recorded program,
f a i l u r e to p r e - t e s t co n ten t fo r g en eral i n t e r e s t , and the
group l i s t e n i n g e f f e c t s open to q u estio n th e h y p o th e sis on
the p r e s t i g e - e f f e e t .
Paulson attem pted to measure some e f f e c t s of the
p r e s tig e of the speaker in causing s h if t- o f - o p in io n on the
to p ic of reducing the minimum voting age to e ig h te e n . He
prepared four recorded speeches, two giving a l l arguments
on "one-side" and two giving the leading argument a g a in st
reducing the age as w ell as the arguments in fav o r of i t .
Paulson p resen ted a p ro fe s s o r and a u n iv e r s i ty sophomore as
sources of high and low c r e d i b i l i t y f o r h i s study. He
t e s t e d speech c l a s s e s w ith both the Woodward b a l l o t and a
m u ltip le -c h o ice r e te n tio n t e s t . Of the men, a s i g n i f i c a n t l y
la rg e r p ro p o rtio n s h if te d t h e i r opinions in the d ir e c tio n of
the "p ro fesso r" than in the d ir e c tio n of the "student"
source. D iffe re n ces in r e te n tio n score between the two
sources of c r e d i b i l i t y f o r both men and women were not sig
n i f i c a n t . The e f f e c t of the p r e s tig e of the speaker on
s h if t- o f - o p in io n was s im ila r to what Haiman had found, w ith
one im portant ex cep tio n . In P au lso n 's study the women did
not s h i f t s i g n i f i c a n t l y more f o r the p re s tig e speaker th an
f o r the nonprestige s p e a k e r .
^ S t a n l e y F. Paulson, "The E ffe c ts of the P re stig e
of the Speaker and Acknowledgement of Opposing Arguments on
Audience R e ten tio n and S h ift of Opinion, Speech Monographs.
XXI (November, 1954), 267-68.
C a th c a rt stu d ied the e f f e c t s of documenting evidence
in the p r e s e n ta tio n of speeches on c a p i t a l punishment when
no inform ation about the speaker was known by the audience.
Twenty-minute recorded speeches were played before stu d en t
au diences, using the Woodward s h if t- o f - o p in io n b a l lo t f o r
measurement of response. C a th c a rt re p o rte d no s u b s t a n tia l
c o r r e l a t i o n between an a u d i t o r 's b e l ie f in the n e c e s s ity of
documenting evidence and h is response to a speech which con-
31
ta in e d no documentation of the evidence p rese n ted . The
m atter of documentation has some bearing on t h i s e x p e ri
m e n te r's study.
No o th er study t h a t has lo g ic a l r e la tio n s h i p to the
measurement of source c r e d i b i l i t y in a mass media has been
discovered by t h i s in v e s tig a to r in the l i t e r a t u r e of e x p e ri
mental re s e a rc h .
L im ita tio n s of p rev io u s stu d ie s
A ll of the experim ental s tu d ie s th a t have been r e
ported in the l i t e r a t u r e and have p u rp o rted to measure the
e f f e c t s of d i f f e r e n t c r e d i b i l i t i e s of speakers have been
e i t h e r s tu d ie s of p latfo rm address or s tu d ie s of s h i f t - o f -
opinion a n a ly s is of p r e s e n ta tio n s t h a t were made by p lay in g
of ta p e -re c o rd e d or tr a n s c rib e d speeches in classroom s e t
t i n g s . Without e x c ep tio n , the stu d ie s have assumed t h a t a
^ R o b e r t S. C a th c a rt, "An Experim ental Study of the
R elativ e E ffe c tiv e n e s s of Four.Methods of P re se n tin g Evi
dence." Speech Monographs. XXII (August, 1955), 228-33.
recorded speech t h a t was played before a sizab le group of
in d iv id u a ls provided a s i t u a t i o n th a t was a p p lic a b le t o an
a c tu a l broadcast s i t u a t i o n . No in v e s tig a to r has designed a
s o u r c e - c r e d ib ilit y experiment th a t was to c o n tro l th e se
fundam ental v a r ia b le s in e v a lu a tin g a ra d io t a l k :
Audience i n t e r e s t and a t t e n t i o n to the in fo rm atio n al
c o n te n t. In h is inform ation film s tu d ie s , Hovland re p o rte d
th a t " i n t e r e s t and the v ario u s ways of arousing i n t e r e s t may
be im portant f a c t o r s to r e l a t e t o ed u c a tio n a l e f f e c t s even
though i n t e r e s t i s not an adequate c r i t e r i o n of e f f e c t i v e -
oo
n e s s." In a ra d io lis te n i n g s i t u a t i o n the audience i s not
" c a p tiv e ." The p r e - t e s t i n g of m a te ria ls fo r a t t e n t i o n and
i n t e r e s t appears t o be e s s e n t i a l to a study of the e f f e c t s
of ra d io l i s t e n i n g . As Harwood and C a r t i e r have expressed
i t :
The p rese n t w r i t e r s b eliev e th e se v a ria b le s — a t t e n
t i o n and i n t e r e s t — are among the basic v a r ia b le s th a t
c o n tro l whether any communication— good or bad— occurs
a t a l l . 33
In d i r e c t referen ce to b ro a d c a stin g , Whan sa id , "Without
i n t e r e s t , no ra d io audience e x i s t s . " 34
^%iovland, Lumsdaine, and S h e f fie ld , op. c i t . . p.
81.
F ra n c is A. C a r t ie r and Kenneth A. Harwood, "Some
Q uestions About A tte n tio n ," The Jo u rn al of Communication.
V III (Autumn, 1958), 108.
34F o re st L. Whan, "The Speech P ro fe ssio n J i l t s
Radio," The Q u a rterly Jo u rn a l of Speech. XXX (December,
1944), 4 4 l.
21
Sponberg re p o rte d as one c o n d itio n of h is experiment
t a l procedure the use of a recorded speech "on a su b ject of
v i t a l i n t e r e s t to the a u d i e n c e ." ^ He developed an argumen
t a t i v e speech on the g en e ra l to p ic of marriage and the war.
However, lik e D i e t r i c h , ^ Haiman, Paulson,^® and High
l a n d e r , ^ Sponberg merely assumed th a t th ere would be spec
i a l i n t e r e s t in the su b je ct m a te ria l. In none of the stud
ie s mention was a p r e te s t made of audience i n t e r e s t in the
inform ation co n ten t of the t a l k used in the experim ental
study. Although Sponberg's study was of a recorded speech,
i t was not lab ele d as a ra d io t a l k . In the o th er s tu d ie s ,
the inform ation was regarded as s u ita b le f o r ra d io p re se n ta
t i o n . No other study was found th a t used a p r e t e s t to ev a l
uate the i n t e r e s t and a t t e n t i o n of the su b je c ts in the ra d io
t a l k t h a t was to be used in the experim ental procedure.
L istening to the rad io ta l k alone or as p a rt of, a
t y p ic a l fam ily grouping. Knower' s pioneer s tu d ie s in the
1930's in d ic a te d th a t th ere are s ig n i f ic a n t p sychological
f a c t o r s a t work in an experim ental s e ttin g when the su b je c ts
hear a speech as p a rt of a group audience as compared to
3^Harold Sponberg, "A Study of the R elativ e E ffe c t
iv en ess of Climax and Anti-Climax Order in an Argumentative
Speech," Speech Monographs. X I I I : 1 (1946), 37.
^ D i e t r i c h , pp. c i t . . p. 58.
37Haiman, op. c i t . . pp. 191-92.
38paulso n , op. c i t . . p. 267.
39H ighlander, op. c i t . . p. 188.
lis te n i n g to a speaker as an audience of one.4* ^ Lindgren
more r e c e n tly emphasized the group e f f e c t on the in d iv id u a l
in an e d u c a tio n a l s e t t i n g . As a p sy c h o lo g ist he p o in te d out
" th a t in d iv id u a ls w ill be a f fe c te d by groups, th a t t h i s in
fluence w i l l be em otional in c h a ra c te r and t h a t a change in
behavior of in d iv id u a ls w ill r e s u l t . " 4 * R a d io 's intim acy as
a medium of communication i s one of i t s prominent c h a ra c te r
i s t i c s . The l i s t e n e r seldom forms p a r t of a lis te n i n g group
A O
la rg e r than the fam ily c i r c l e and he o ften l i s t e n s alone.
Yet t h i s in v e s tig a to r found no re p o rt of rese arch t h a t took
t h i s intim acy in to account in t e s t i n g the l i s t e n e r s t o ra d io
t a l k s f o r r e te n tio n of inform ation or change of a t t i t u d e s .
L im itatio n s in c o n tro llin g the p h y sic a l v a r ia b le s of
playing a recorded program to a classroom of stu d e n ts and
attem pting to apply t h e i r audience r e a c tio n s to normal ra d io
l i s t e n i n g . T h is in v e s tig a to r found no study th a t did not
use the classroom or the m ilita r y o r ie n ta tio n le c tu re h a l l
A Q
as the s ite of lis t e n i n g . V a ria b le s of lis te n i n g under
40F ra n k lin H. Knower, "Experimental S tu d ie s of
Changes in A ttitu d e s : 1. A study of the E ffe c t of O ral Argu
ment on Changes of A ttitu d e ," Jo u rn a l of S o c ia l Psychology.
VI (August, 1935), 343.
4 *Henry C. Lindgren, "The E ffe c t of the Group on the
Behavior of the In d iv id u a l," E ducation. LXXIII (February,
1953), 387.
4^Edgar E. W i llis , Foundations in B roadcasting (New
York: Oxford U n iv e rs ity P res s * I n c . , 1951J, pp. 118-19.
43
Hovland, Lumsdaine, and S h e ffie ld , op. c i t . . p.208.
23
these c o n d itio n s include group o ra l r e a c tio n s , proxim ity of
the in d iv id u a l to the audio speaker, tem perature and accous-
t i c d i f f e r e n t i a l s in the v ario u s classroom s, v a r ia tio n s in
the time of day when the groups were t e s t e d , and the g en e ra l
e f f e c t of being a "cap tiv e" audience in a ty p ic a l classroom
s i t u a t i o n . ^ T his experim enter attem pted t o improve upon
the c o n tro l of these v a r ia b le s .
Summary
D espite the sig n ific a n c e of the broadcast media in
the U nited S ta te s , most of our knowledge of the e f f e c t s of
ra d io in the dissem in atio n of inform ation has been based
upon co n c lu sio n s drawn from nonexperim ental s tu d ie s . The
l i t e r a t u r e r e v e a ls l i t t l e rese arch a n a ly s is of the v a r ia b le s
th a t make one in fo rm atio n al program more im pressive than
an o th er.
P ro fe s s io n a l b ro a d c a ste rs and ed u cato rs have d i f
fered in t h e i r views of the importance of the speaker in the
p re s e n ta tio n of the inform ative ra d io t a l k . Some have con
tended th a t the i n t e r e s t of the l i s t e n e r in the co n ten t i s
most im portant, b eliev in g t h a t if th e re i s i n t e r e s t in the
to p ic and the m a te ria l i s we11-prepared, the voice of the
44
Howard W . Townsend, "F actors of Influence in
Radio Speech," The Q u a rte rly Jo u rn al of Speech. XXX (A p ril,
1 9 4 4 ),1 8 8 .
......................................................................................' 24
p ro fe ssio n a l rad io announcer w i l l impart as much knowledge
as the voice of the s p e c i a l i s t . O thers have in s is te d th a t
the voice of the in fo rm a tio n -s p e c ia lis t adds p re s tig e to
the topic and produces more a t te n t io n and g re a te r under
standing than the voice of the announcer. No experim ental
evidence has been found to support co n c lu siv e ly e i t h e r point
of view.
I t was the purpose of t h i s study to provide informa
t io n with which b ro ad ca ste rs and educators could improve
t h e i r uses of ra d io t a l k s as a means of imparting informa
t io n to the p u b lic .
I t was the problem of t h i s study to measure the
e f f e c t s of i n t e r e s t in the to p ic and the eth o s of the speak
e r upon understanding of an inform ative ra d io ta lk .
This in v e s tig a to r assumed th a t the la b o ra to ry se t
tin g used in t h i s experiment re s u lte d in audience measure
ments c lo s e ly approximating the normal responses of the sub
j e c t s when they were tuned to the ra d io b roadcasts in the
home. I t was also assumed th a t the matched groups were rep
re s e n ta tiv e of the public in g en eral.
Examination of the l i t e r a t u r e revealed no e x p e ri
mental study d i r e c t l y r e la te d to the problem of this; inves-
i
t i g a t o r . Some in d ire c t r e la tio n s h ip s to t h i s in v e s tig a tio n
were discovered in the s h ift- o f- o p in io n stu d ie s of source
c r e d i b i l i t y .
The few experim ental stu d ie s th a t were intended to
measure l i s t e n e r responses to ra d io t a l k s used recorded
speeches t h a t were played t o large groups of s u b je c ts . The
t a l k s were not p r e - te s t e d fo r in d ic a tio n s th a t the l i s t e n e r
would a c tu a l ly s e le c t such a t a l k on the b a s is of i n t e r e s t
i f he were not a " c a p tiv e ” audience. T his in v e s tig a to r a t
tempted to c re a te an experim ental procedure th a t would avoid
the l i m i ta tio n s in d ic a te d h ere.
Preview of Remaining C hapters
C hapter II is devoted to d e ta ile d e x p lan atio n of the
methods, te c h n iq u e s, and procedures used in the experim ent,
including the p re p a ra tio n of the ra d io t a l k , the s e le c tio n
of the re a d e r, c o n s tru c tio n of the measuring in stru m en ts,
and the conduct of the experiment in the rad io stu d io .
C hapter I I I is concerned w ith the p re s e n ta tio n and
i n t e r p r e t a t i o n of the d a ta . S t a t i s t i c a l trea tm en t of the
data r e l a t i n g t o the in v e s tig a tio n of seven n u ll hypotheses
i s re p o rte d w ith in t e r p r e t a t i o n of the r e s u l t s .
C hapter IV in c lu d es a summary of the study, conclu
sions based upon the i n t e r p r e t a t i o n of data rep o rted in
C hapter I I I , im p lic a tio n s t o be drawn from the r e s u l t s of
the study, and suggestions fo r f u r t h e r re se a rc h .
CHAPTER I I
METHODS, TECHNIQUES,
AND PROCEDURES
Methods
D eterm ination of the method
of re searc h
The purpose of t h i s study re q u ire d th a t the e x p e ri
mental method of re se a rc h be used. The p relim in ary e x p e ri
mental design appeared to meet the basic c o n s id e ra tio n s
suggested by Whitney f o r the proper use of the experim ental
method:
At l e a s t seven basic c o n s id e ra tio n s should guide:
c a r e f u l d e f in it io n of an im portant, f e a s ib le problem;
minute d e f i n i t i o n of the experim ental f a c to rs ; s e le c
t io n of the best te ch n iq u es fo r the experim ental setup;
c o n tro l of a l l f a c t o r s in accord w ith the law of the
sin g le v a ria b le ; a d m in is tra tio n of an adequate t e s t i n g
program; a d e ta ile d readable re p o rt of re se a rc h e x p e ri
ences and r e s u l t s ; and d e f in it e p ro v isio n f o r the repe
t i t i o n of the experiment and f o r f u r th e r re se a rc h in
the f i e l d of in q u iry .
| ^Frederick Lamson Whitney, The Elements of Re
se a rc h (3rd e d .: New York: Prentice-H a 11. I n c ., 1950), p.
j a is:~
26
Techniques
S e le c tio n of the topic f o r the
ra d io t a l k
T h is experim enter re q u ire d a topic th at, would meet
fiv e e s s e n t i a l requirem ents fo r purposes of t h i s study:
1. Both men and women should have general personal
i n t e r e s t in the inform ation to be p resented.
2. The to p ic should be somewhat f a m ilia r to the
s u b je c ts , and should be one on which they had
l i t t l e sp e c ific knowledge but some opinion.
3. The to p ic should be ap p ro p riate as program mat
e r i a l fo r broadcasting as a short fe a tu re t a l k
in a longer ra d io program.
4. The to p ic should lend i t s e l f to a proper t e s t
of r e te n tio n of inform ation.
5. The to p ic should be acceptable as a statem ent of
an ex p ert or as th a t of a person who i s a s k ille d
re a d e r.
H ealth and medical asp ects of the topic of sleep
were se lec ted fo r the study. T his kind of inform ation ap
peared to meet the requirem ents of the in v e s tig a to r . Sleep
i s of i n t e r e s t to both men and women. D espite the impor
tance of sleep to h e a lth , public knowledge of the rese arch
on sleep i s s l i g h t . Most in d iv id u a ls have some opinions on
t h i s to p ic . In v e s tig a tio n revealed th a t a s u b s ta n tia l
amount of inform ation on sleep was available*
28
T his in v e s tig a to r spent sev eral months in the com
p i l a t i o n of d e ta ile d inform ation on the topic of sleep .
From th e se d a ta , f o r ty items of inform ation were se lec ted
fo r in c o rp o ratio n in to the rad io t a l k . This inform ation was
examined by fiv e s ta f f members of the Speech Department of
Oregon S tate C ollege, and t h i r t y - f i v e items f i n a l l y were se
le cte d fo r use in preparing the ra d io s c r i p t . Two p ro fe s
sio n a l rad io w r ite r s of rad io s ta tio n KOAC, the sta te educa
ti o n a l rad io s ta tio n of Oregon, were consulted on the prepa
r a tio n of the m a te ria l and on the maximum length of the ra
dio t a l k th a t should be incorporated in to the program format
th a t was to be used in the study. Because the ta lk was to
be a fe a tu re of a "Family Hour" program, the length of the
t a l k was lim ite d to a maximum of e ig h t minutes of d e liv e ry
tim e .
In preparing the rad io s c r ip t , sp e c ia l co n sid era tio n
was given to c o n v e rsatio n a l s ty l e , dramatic i n t e r e s t , sim
p l i c i t y of language, and v a r ie ty of sentence s tr u c tu re .
A fter the f i r s t ra d io t a l k was w ritte n , a second v ersio n was
prepared; the second v ersio n contained the same basic in
form ation th a t was in the f i r s t s c rip t but several changes
were made in the arrangement of inform ation. Sources of the
inform ation were not sta te d in the s c r ip t . T his in v e stig a
to r believed th a t documentation of the content would add to
the study a v ariab le th a t could not be adequately c o n tro lle d .
A s t a f f member of Oregon S tate College who had pro
fe s s io n a l announcing experience was selected to d e liv e r the
two radio t a l k s . Both t a lk s were recorded in the radio
studio in Shepard H all on the campus of Oregon State Col
lege. The recording engineer used a studio dynamic micro
phone, E lectro Voice Model No. 6b4 fo r both p re se n ta tio n s.
Both ta lk s were recorded upon Scotch brand p la s tic recording
tape No. I l l A-12 a t a tape speed of seven and one-half
inches per second. The engineer attempted to hold the vol
ume le v e ls constant as the ta lk s were recorded through a
Gates audio console into a Magnecorder Model PT6-AH. This
model of the Magnecorder i s widely used in commercial radio
s ta tio n s f o r making p ro fe ssio n a l tape recordings.
After the two radio ta lk s were recorded, t h i s inves
t ig a to r te s te d the ta lk s on fo rty -e ig h t college students who
were drawn from the basic speech c la s s e s a t Oregon State
College. The students in groups of six were in v ite d to the
Shepard H all radio studio to take p a rt in the experiment.
At the studio, one of the two recorded t a lk s was played to
each group with no in tro d u ctio n or prelim inary explanation;
The tape recordings were played from the studio c o n tro l room
through the Magnecorder and the Gates audio console, and the
su b jects heard the ta lk s through a ten -in c h Meissner speaker
th a t was b u i l t in to the studio w all adjacent to the co n tro l
room.
A fter the t a l k had been presented, each student was
given the following b a llo t fo r ev a lu atio n of in te r e s t:
3 0
Please in d ic a te by an X in the proper p o s itio n ,
your candid opinion of the t a l k th a t you have j u s t
heard.
I found the t a l k (A)— in te re s tin g (B )--highly
in te r e s tin g (C)— d u ll.
Thank you fo r your cooperation.
Twenty-four males and tw enty-four females p a r t i c i
pated in t h i s p a rt of the study. One group of twelve males
and twelve females lis te n e d to Speech #1; the remainder of
the su b je cts heard Speech #2. A ll su b jects described the
two t a l k s as in te r e s tin g or highly in te r e s ti n g . As shown in
Table 1, Speech #2 received a higher r a tin g than Speech #1,
Table 1
INTEREST RATING OF THE TW O RECORDED TALKS
T alks I n te r e s tin g _ I n te r e s tin g DuU
Speech #1 18 . 6 0
Speech #2 15 9 0
with 9 students ra tin g the speech as highly in te r e s tin g .
Only 6 of the students who heard Speech #1 in d icated t h i s
t a l k was highly in te r e s ti n g .
The in v e s tig a to r then made minor re v is io n s in sen
tence s tru c tu re of Speech #2. Speech #2 was again recorded
in the same voice th a t produced the other p re s e n ta tio n s .
31
This tape recording was made on the same equipment th a t was
described above. T h irty stu d en ts were se le c te d from the
basic speech c la s s e s a t Oregon S ta te College to s ta te t h e i r
in te r e s t in the rev ised rad io t a l k . None of these students
had been used in the p r io r experim ent. F ifte e n male sub
je c ts and f i f t e e n female su b je cts were brought to the Shep
ard H all rad io studio in groups of six f o r the e v a lu a tio n .
The same ra tin g scale was used and th e re was no prelim in ary
explanation of the experim ent. Seventeen su b je c ts marked
the t a l k as " in te re s tin g " and th i r t e e n su b je cts marked the
ta lk as "highly in te r e s ti n g ." No su b ject in d ic a te d th a t the
ta lk was d u ll.
In t h i s prelim inary te s tin g the rad io t a l k appeared
to be of in t e r e s t to the prospective kinds of l i s t e n e r s .
Later a d d itio n a l te s tin g su b sta n tia te d t h i s observation.
The rev ised Speech #2 was used without change in a l l f u r th e r
te s tin g procedures.
S ele ctio n of the voice to
rep rese n t the two types of
a u th o rity
This study req u ire d a vocal d e liv e ry th a t would be
equally acceptable to the l i s t e n e r as th a t of a reputable
physician or th a t of a news commentator in a large metropol
ita n ra d io s ta tio n . On the bases of d i s t i n c t i v e voice char
a c t e r i s t i c s and p ro fe ssio n a l background in ra d io announcing,
three persons were chosen fo r t h i s p a rt of the study. A ll
32
were on the s ta ff of Oregon State C ollege. One was a member
of the Speech Department; the other two were radio announc
e rs on the s ta ff of KOAC, the state educational s ta tio n of
Oregon th a t had studios on the campus of the co lleg e. The
revised radio t a l k was tape-recorded by the three s ta ff mem
bers in the Shepard H all rad io studio, using the recording
equipment th a t was described e a r l i e r in t h i s chapter. Each
recording was made a t a speed of seven and one-ha If inches
per second, and the th ree announcers completed t h e i r ta lk s
w ithin the suggested lim ita tio n of eig h t minutes.
Ninety college students who were drawn from the
basic speech c la s s e s p a r tic ip a te d in a t e s t of the three
voices. F o rty -fiv e men and f o rty -f iv e women were chosen.
A ll subjects were eighteen or nineteen years of age and were
c l a s s i f i e d by the r e g i s t r a r 's office as freshmen or sopho
mores. The ninety students were divided in to groups th a t
numbered five to eig h t subjects; these groups were sched
uled to be in the Shepard H all radio studio during morning
hours when the studio was av a ila b le to the experiment.
T h irty students lis te n e d to the recorded radio ta l k as de
liv e re d by Speaker A, t h i r t y heard Speaker B, and the re
maining t h i r t y lis te n e d to Speaker C. Each group of sub
j e c ts fo r t h i s phase of the experiment was composed of f i f
teen males and f i f t e e n fem ales. No in tro d u ctio n was made to
the playing of the recorded speech. The following s ta te
ments were handed to each subject in the studio before the
33
recorded p re se n ta tio n :
Your cooperation i s requested to a s s i s t in a research
p ro je c t involving about te n minutes of your tim e. You
need not put your name on t h i s sheet. You are asked to
l i s t e n to a short rad io ta l k which was presented as a
p a rt of a longer rad io program over a large m etropolitan
rad io s ta tio n . A fter lis te n in g to the t a l k , you are
asked to s e le c t the person who d eliv e red the t a l k from
the two p o s s i b i l i t i e s l i s t e d below. Please mark the one
you f e e l , using your own personal lis te n in g c r i t e r i a .
C irc le the appropriate choice.
1. This t a l k was d eliv ered by Chet P arker, rad io
news an aly st and commentator fo r a large m etropoli- .
ta n rad io s ta tio n .
2. This t a l k was d e liv ered by Dr. Ross Maloney,
M.D., Physician in Research fo r a large m etropolitan
h o s p ita l.
Now please mark one of the choices below in the
space in d ic ated :
I found t h i s ta lk : i n t e r e s t i n g , Highly I n te r
e s tin g , . Dull.
Table 2 p rese n ts the data r e la tin g to the i d e n t i f i
c a tio n of the speaker and the degree of in t e r e s t in the re
corded t a l k . The su b jects were almost eq u ally divided in
t h e i r i d e n tif ic a tio n of the voice used fo r Speech #1 as
rep rese n tin g e i th e r the physician, Dr. Maloney, or the news
34
TABLE 2
SPEAKER IDENTIFICATION AND INTEREST RATING
OF THE RECORDED TALKS
Recorded
Talk
Parker Maloney In te re stin g
Highly
In te re stin g
Dull
Voice #1 14 16 20 10 0
Voice #2 20 10 23 7 0
Voice #3 18 12 21
9 0
commentator, Chet Parker. The voices used fo r speeches #2
and #3 appeared to have more c h a r a c te r i s tic s th a t were asso
c ia te d with a news commentator than with a physician. Per
sonal interview with te n of the students who heard Speech #1
revealed th a t these students believed th a t e i th e r i d e n t i f i
c a tio n of the speaker would be acceptable to them.
The selected recording, Voice #1, was fu rth e r te s te d
under co n tro lled conditions sim ila r to those th a t were re
ported fo r the previous experiment. A group of f i f t e e n men
and f i f t e e n women was chosen from the basic speech courses
at Oregon S tate C ollege. Sixteen students were eighteen
years of age and fo u rteen students were nineteen years of
age. The su b jects were te s te d in groups of fiv e th a t were
segregated by sex; each group was scheduled to appear in the
Shepard Hall radio studio during the morning hours of e i th e r
10:00 A.M. or 11:00 A.M. Before lis te n in g to the recorded
ta lk , each subject was given copies of the same statem ents
35'
th a t were used in the previous experiment. Male su b je cts
and female su b jects were almost eq u ally divided in id e n ti
fying the speaker as e i t h e r Dr. Maloney, the p h y sician , or
Chet P ark er, the newsman. Table 3 also showed l i t t l e sex-
d iffe re n ce in the i n t e r e s t r a tin g s . A ll su b je cts reported
Table 3
SPEAKER IDENTIFICATION AND INTEREST RATING
OF THE SELECTED TALK
Group Parker Maloney Intere stin g
Highly
In tere stin g
Dull
M a le s . 9 6 8 7 0
Female s 7 8 9 6 0
T o ta l 16 14 17 13 0
th a t they found the t a l k e i t h e r in te re s tin g or highly in te r -
e s t i n g .
A fter interview ing th ese stu d e n ts, t h i s in v e s tig a to r
had reasonable assurance th a t the voice on the se lec ted re
cording would be accepted by the l i s t e n e r if introduced as
e i th e r Dr. Maloney or Chet Parker. In a l l of the prelim
inary te s tin g no request was made fo r the students to iden
t i f y themselves by w ritin g t h e i r names on the t e s t s ; the
purpose of the anonymity was to have strong assurance th a t
the measurements were of candid expressions of the personal
opinions of the student su b jects.
36
P re p aratio n of the re te n tio n
t e s t
Experience in the te s tin g of college students and a
review of l i t e r a t u r e convinced t h i s in v e s tig a to r th a t pre
te s tin g the college su b je cts on what they knew of the topic
of sleep would create a psychological v aria b le th a t should
be avoided. P arish and Campbell commented upon a hazard of
p r e - te s tin g :
The t r a d i t i o n a l p r e te s t and p o s tte s t experim ental
and c o n tro l groups design has one weakness which has
been g en e rally overlooked. S t r i c t l y speaking, as a
lo g ic a l model i t provides a b a s is of g e n e ra liz a tio n only
to other p re te s te d popu latio n s. This lim ita tio n , t r i v
i a l in many experim ents, may be expected to be important
in a t titu d e change s tu d ie s . The p r e te s t must in e v ita b ly
cre a te a set or focus a t te n t io n on the topic under
study, making the p re te s te d audience no longer ty p ic a l
of audiences in g en e ra l. I t likew ise w ill in many s it u
a tio n s a l e r t the audience to the a tte n tio n s of the ex
perim enter and propaganda. For those re se a rc h e rs i n t e r
e ste d in g en eralizin g to the u n p retested audience, i t
would seem w ell to avoid the p r e te s t and p o s tte s t de
sign and to s u b s titu te fo r i t a design in which equiva
le n t groups receive a single t e s t , preceded f o r the ex
perim ental group by exposure to the experim ental v a ri
able.
Schank and Goodman,^ and Hovland^ have expressed
o
Jack A. P a rris h and Donald T. Campbell, "Measuring
Propaganda E ffe c ts with D irect and In d ire c t A ttitu d e T e s ts ,"
The Jo u rn al of Abnormal and S o cia l Psychology. XLVIII
(January, 1953J , 4.
3
R. L. Schank and C. Goodman, "Reactions to Propa
ganda on Both Sides of a C o n tro v e rs ia l. Issu e ," Public Opin
ion Q u arterly (No. 3, 1939), pp. 107-12.
4C a rl I. Hovland, Arthur Lumsdaine, and Fred Shef
f i e l d , Experiments on Mass Communication. Vol. I l l : S tudies
in S o cial Psychology in World War II (4 v o l s . : Princeton:
P rinceton U n iv e rsity Pre s s , 1949-bO), p . 28.
37
sim ila r views on a t t i t u d i n a l stu d ies. A p o s t- te s t alone was
used in the present study.
The m ultiple-choice question appeared to have ad
vantage over the t r u e - f a l s e item f o r te s t i n g the amount of
inform ation th a t was re ta in e d and sim ultaneously providing
freedom to r e je c t inform ation th a t was co n trary to general
opinion.
Cronbach described six re sp o n se -se ts in fiv e common
types of t e s t s and declared:
The m ultiple-choice item, with the subject forced to
choose on every item, i s the only form in common use f o r
pen cil-an d -p ap er t e s t s which appears to be fre e from
response s e ts . This i s one reason why i t is being in
c re a sin g ly used in t e s t s pf knowledge, a t t i t u d e s , in
t e r e s t s , and p e rs o n a lity .
This in v e s tig a to r l i s t e d t h i r t y - f i v e items of in fo r
mation th a t had been incorporated in to the radio t a l k and
prepared m ultiple-choice questions based on t h i s inform ation
fo r use as a t e s t of r e te n tio n of inform ation. In s e le c tin g
the questions c o n sid e ra tio n was given to the probable d i f f i
c u lty of the t e s t item, the p o s s i b i l i t y of preparing a mul
tip le -c h o ic e item with eq u ally strong choices, and the in
clu sio n of several q u estio n s th a t would r e f l e c t the sub
j e c t ' s psychological a t t i t u d e toward the speaker. T h irty -
five t e s t items were co n stru cted ; these were then evaluated
by members of the departments of Speech and Psychology at
5
Lee J . Cronbach, E s s e n tia ls of Psychological T est
ing (New York: Harper & B ro th e rs, 1949f, p. b l.
38
Oregon S ta te C ollege. On the b a sis of recommendations by
the fa c u lty members, t h i s experim enter then revised the
questions and se le c te d tw enty-five m ultiple-choice items fo r
prelim inary t e s t i n g .
These tw enty-five questions then were te s te d with
eighty su b jects who were drawn from the c la s s e s in basic
speech a t the c o lle g e . The purpose of t h i s t e s t was to de
termine the scores th a t might be expected on the b a s is of
p r io r knowledge and chance. Forty male su b je cts and f o rty
female su b je cts were chosen fo r t h i s p a rt of the study.
These su b je cts were selec ted because they were reasonably
re p re se n ta tiv e of the college population from which other
su b jects would l a t e r be drawn. The eig h ty su b jects were a l
most e q u a lly divided between freshmen and sophomores, and
a l l were e i t h e r eig h teen or nineteen years of age.
The t e s t was adm inistered in two classrooms in Shep
ard H all. The su b je cts were segregated by sex; they took
the t e s t in groups ranging in size from f i f t e e n to twenty
su b jects fo r each te s tin g session. Four s ta f f members as
s is te d in supervising the t e s t . All te s tin g was done during
a period of fiv e days between the hours of four and five
P.M.
The te ach e r in charge of each room d is tr ib u te d the
t e s t and then read the follow ing in s tru c tio n s :
This t e s t is p a rt of an experim ent. Please tr y
to answer a l l of the questions on the basis of your
present knowledge and your personal opinion. Try to
make some kind of response to every question. You
need not place your name on your paper. Your coop
e ra tio n on t h i s experiment is deeply appreciated.
These in s tru c tio n s were also included on the t e s t
paper. A copy of the in s tru c tio n s and the t e s t is included
in the Appendix. A maximum time of twelve minutes was a l
lowed fo r the completion of the t e s t . All subjects com
pleted the t e s t e a s ily w ithin the a l lo tte d tim e. A student
a s s is ta n t then c o lle c te d the papers, inspecting each one to
make c e rta in th a t answers to a l l the items on the t e s t had
been attempted by each student.
A fter a l l the papers had been assembled, they were
graded on the b a sis of one point fo r each question, with a
p e rfe c t score represented by tw enty-five p o in ts. On the
b a sis of p rio r knowledge and chance, no score above th ir te e n
was recorded by any of the eig h ty subjects te s te d . From
examination of Table 4 there appeared to be a s u b s ta n tia l
amount of inform ation on the topic of sleep th a t was unfamil
ia r to the su b jects who took the t e s t . The mean score on
the t e s t was 9.25 fo r the women as compared to 8.9b fo r the
men. The mean score of a l l students who took the examina
tio n was 9.10.
The women who took the t e s t appeared to have a
s lig h tly g re a te r knowledge of the inform ation than did the
men on the b asis of p rio r knowledge and chance. The d if fe r-
40
TABLE 4
SCORES ON TEST BASED ON PRIOR KNOW LEDGE AND
CHANCE
Subjects Mean Score Range of Scores
40 Men 8.95 5 to 14
40 Women
9.2b 6 to 14
Both Groups 9.10 5 to 14
ence in mean score was not regarded as s u f f ic ie n tly d is
crim inatory fo r making the t e s t unsuitable f o r the purposes
of t h i s study. Observation of the raw data showed a good
d is tr ib u tio n in the t e s t grades of wrong choices on the
trip le -c h o ic e d questions.
Because of the length of the t e s t , the individual
items required s u ffic ie n t in d ices of d i f f i c u l t y to provide
v alid d isc rim in a tio n between matched groups of students who
were su b jects in the main p art of the experiment. The d if
f i c u l t y index of each item was computed. This index is ob
tain ed by to ta lin g the number of rig h t answers to a given
item, and then dividing t h i s sum by the number of answers
rig h t or wrong. This index ranges from 0 to 1. If a l l
answers are r i g h t , the index is 1. As recorded in Table 5
under heading P, the d i f f i c u l t y index of the individual t e s t
items ranged from .091 to .738.
The d i f f i c u l t y index was then co rrected fo r a
41!
TABLE 5
ITEM DIFFICULTY ON BASIS OF PRIOR KNOW LEDGE
AND CHANCE
Item No. No. of People
,, T , Corrected
D iffic u lty Index D if fic u lty
Inde x* Pc
1 80 .300 -.050
2
80 .475 .212
3 80 .538 .306
4 80 .500 .250
5 80 .600 .400
6 80 .262
-.106
7 80 .162
-.256
8 80 .150 -.275
9 79 .595 .392
10 79
.152 -.272
11 80 .350 .025
12 77 .091 -.364
13 80 .650 .475
14 80 .388 .081
15 79
.202 -.196
16 78 .474 .212
17 80 .275 -.088
18 80 .175 -.238
19 80 .325 -.0 1 2
20 80 .238 -.144
21 80
.562 (two-choice item)
22
80 .738 .606
23 80 .250 - . 125
24 80 .338 .006
25 80 .338 .006
*The corrected d if f ic u Ity inde x
R-1/2W
fo r a trip le -c h o ic e d
re sponse is defined as Pc=- N with the index ranging
from - .5 to 1. If a l l answers are rig h t., the index is equal
to 1.
42
R - 1/2W
trip le -c h o ic e d question. The formula Pc = ----- ^------ was used
in which R re p re se n ts the number of rig h t answers, W the
number of wrong answers, and N the number of subjects who
answered the sp ecific question. This index ranges from - .5
to 1. If a l l answers are r i g h t , the index is equal to 1.
If a l l answers to a sp ecific question are wrong the index is
equal to - . 5 . In Table 5 the co rrected d i f f i c u l t y index fo r
each item is recorded in column Pc.
The co rrected d i f f i c u l t y index ranged from .606 to
-.3 6 4 , with only two questions showing a d i f f i c u l t y index
above .400. The range of d i f f i c u l t y fo r the remainder of
the items appeared s u f f ic ie n tly strong to j u s t i f y t h e i r in
clu sio n . The two items with poor d i f f i c u l t y indices were
re ta in e d to add length to the te s tin g instrum ent. Although
item #21 was not adaptable to a three-choiced response, i t
was reta in ed because of i t s p o te n tia l fo r measuring both in
formation and a t t i t u d e . As a two-choice question, i t had an
ex c ellen t d i f f i c u l t y index of .562 as compared to a p erfect
index score of .500.
Because the te s tin g instrument was to be given to
matched groups of su b jects, i t was necessary fo r t h i s ex
perimenter t o measure the v a l id ity of matching the subjects
fo r in te llig e n c e by taking the grade-point average of the
previous term as the index of in te llig e n c e .
If the t e s t were v a lid , the students with the higher
GPA averages should make higher scores on the re te n tio n t e s t
43
than those su b je cts who had the lower GPA averages. In or
der to determine the v a l id ity of the t e s t , the in v e s tig a to r
decided to t e s t groups of su b jects who would be exposed to
the recorded radio ta l k without the v aria b le of speaker-
source.
The su b je cts were drawn from the basic speech
c la s s e s as being ty p ic a l of su b je cts who were to p a r tic ip a t e
in the f i n a l p a rt of the experim ent. With the cooperation
of the Speech Department of Oregon State C ollege, seven
c la s s e s of students were te s te d in t h i s p a rt of the study.
The c l a s s enrollm ents ranged in size from fo u rteen to twenty
stu d en ts. Seventy-two male su b je c ts and f i f t y - t h r e e female
su b je cts were e n ro lle d in the c la s s e s .
The follow ing announcement was tap e-reco rd ed by a
s ta ff member of the Speech Department and spliced to the
tape of the rad io ta lk :
You are about to hear a short ra d io t a l k . A fter
you have heard the t a l k , you w ill be asked to in d i
cate your personal re a c tio n s to t h i s t a l k . You w ill
be asked to r a te the t a l k as in te re s tin g to you,
highly in te r e s tin g , or d u ll. The ev a lu atio n mater
i a l w ill be given to you a f t e r you hear the t a l k .
You need not place your name on the m a te ria ls th a t
w ill be given to you.
This experim enter arranged to play the t a l k during
the f i r s t p a rt of a re g u la r c l a s s session f o r each of the
44
groups to be te s te d . Before the beginning of the designated
c la ss period, a portable Magnacorder, Model #FT6-AH, was
in s ta lle d in the classroom and the recording was placed on
the tape recorder. After the c la s s had assembled, the re
corder was turned on without a d d itio n a l explanation. A fter
the recording had been played the subjects were given the
twenty-five items, the th re e -p o in t ra tin g scale of in t e r e s t ,
and an a d d itio n a l page on which the students were asked to
indicate age, sex, year in co lleg e, and grade-point average
of the previous term. The playing of the ta lk and the com
p letio n of the t e s t averaged approximately tw enty-four min
utes fo r the seven c la s s e s . Although the c la s s enrollm ents
to ta le d 12b, only 120 students were in attendance on the
days of th i s experiment. Of the 120 su b je cts, seventy were
men and f i f t y were women.
The papers were then evaluated by t h i s experim enter.
In te re s t in the information remained c o n siste n t with previ
ous t e s t r e s u lt s . Seventy-four subjects found the ta lk in
te re s tin g and f o r ty - s ix subjects indicated th a t the ta lk was
highly in te re s tin g . The scores on the t e s t ranged from a
high score of 24 to a low score of 13. The mean score of
the group was 19.63. There were th ir te e n e rro rs of omission
involving ten t e s t items in the t o t a l group of 120 papers.
Errors of omission were scored as in c o rrec t items because a
co rre ctio n formula would have had no sig n ific a n t e f fe c t on
the mean score of the group. In the t h i r t y - s i x papers to
be given a d d itio n a l s t a t i s t i c a l treatm ent there were no e r
ro rs of omission to be computed. The top 18 scores and the
bottom 18 scores on the r e te n tio n t e s t were used to d e te r
mine whether the students who rep o rte d high grade-point av
erages of the previous te rm 's stu d ie s made s ig n if ic a n tly
b e tte r scores on the r e te n tio n t e s t than the su b je cts who
reported low GPA r a tin g s . The follow ing null hypothesis was
te s te d :
Both the h ig h est scores and the lowest scores
on the r e te n tio n t e s t were made by su b je cts who had
the same average in te llig e n c e as measured by the
in d iv id u a l s u b je c t's grade-point average of the pre
vious term.
The t - t e s t was used with a n a ly s is of v aria n ce. As
shown in Table 6, the value of t is 6.93 with 34 degrees of
freedom. This t value is s ig n if ic a n t a t the 1/& le v e l of
TABLE 6
RELATIONSHIP OF MATCHING BY GPA TO SCORES
ON RETENTION TEST
Source of
V a ria tio n
Sum
of Square s
Degrees
of Freedom
Mean
Square
F
Top Scores
vs. Bottom
1 8.3425 48.0559
Within Group 5.9020 34 1736
T o tal 14.2445 35 t = 6.932236*
*This t-v a lu e i s s ig n ific a n t a t the 1% le v e l.
46
confidence. I t may be concluded th a t the hypothesis is
re je c te d , and th a t the su b je cts who made the h ig h e st scores
on the r e te n tio n t e s t had higher GPA's th an those students
who made the lowest scores on the t e s t . From t h i s informa
tio n i t appeared reasonable to conclude t h a t matching on the
b a s is of previous-term GPA would be j u s t i f i e d .
P relim inary te s tin g showed th a t the radio t a l k was
of general in te r e s t to the su b je cts who heard i t . The
speaker apparently would be acceptable to the l i s t e n e r if
introduced as e i t h e r a physician or a news commentator. The
d i f f i c u l t y index fo r the t e s t items demonstrated th a t th ere
would be s u f fic ie n t score d if f e r e n t i a t i o n s f o r ev alu atin g
matched groups of students using the p aire d t - t e s t . Match
ing f o r in te llig e n c e on the b asis of previous-term GPA ap
peared j u s t i f i e d .
Incorporating the rad io t a l k
into a longer program format
S u i t a b i l i t y of inform ation and length of t a l k were
considered c a re fu lly when the rad io t a l k was prepared as
p a rt of a longer rad io program known as the "Family Hour."
A fter c o n s u lta tio n with members of the s ta f f of the Speech
Department and the production s ta ff of KOAC, two introduc
tio n s were se le c te d fo r use in the experim ent. Each in tr o
duction was preceded by the f i n a l twenty seconds of a re
corded piano arrangement of " I ' l l Follow My Secret Heart"
from the RCA V ictor Album, Piano R everies. No. P-64.
47
Both of the in tro d u c tio n s were recorded by a pro
f e s s io n a l s ta ff rad io announcer of KOAC, the state-owned ed
u c a tio n a l s ta tio n of Oregon. For those su b jects who were to
l i s t e n to the t a l k as c re d ite d to Chet P arker, the follow ing
in tro d u c tio n was recorded:
And now i t ' s time fo r the Family Hour Feature
Page, with our KLZ news a n a ly st and commentator,
Chet P arker. Today our newsman r e p o r ts on a subject
close t o a l l our h e a rts . Here is Chet Parker.
For the in tro d u ctio n of the t a l k as ascrib ed to the
p hysician, Dr. Maloney, the follow ing announcement was r e
corded:
I t ' s Feature Time on the Family Hour. Today's
guest i s Dr. Ross Maloney, P hysician in Research fo r
the Denver Municipal H o sp ita l. Here is Dr. Ross
Maloney with a sp e c ia l rep o rt on a subject clo se to
a l l our h e a rts . Dr. Maloney.
Both in tro d u c tio n s were recorded on tape in the
Shepard H all rad io studio.
To provide the l i s t e n e r s with an impression of a
c o n tin u ity of program m a te ria l, the opening passages of a
piano arrangement of the popular song, "Easy t o Love" were
tape-recorded and then slowly faded out. This se ctio n of
tape was then sp liced to the end of the recorded ra d io t a l k .
The song was obtained from the RCA V icto r Album, Piano Rev
e r i e s . No. P-64. Both piano s e le c tio n s used fo r t h i s study
48
were played on a Gates tu rn ta b le a t the proper speed of
seventy-eight rev o lu tio n s per minute and recorded on the
Magnecorder in the Shepard H all studio at the same speed
and audio le v el as had been used fo r the recording of the
radio ta lk .
Tape-recorded in s tru c tio n s fo r
use in the experiment
A member of the s ta f f of the Speech Department re
corded the following in s tru c tio n s fo r use in the experiment:
You are about to hear a short radio ta lk pre
sented as p a rt of a longer program by a lo c al radio
s ta tio n in Denver, Colorado. This was not a network
broadcast, but was tape-recorded and made av ailab le
fo r our use in t h i s experiment. A fter we present
t h i s ta lk , you w ill be asked to give your re a c tio n s
to the ta lk . W e sh a ll ask you to mark a b a llo t in
d icatin g whether you p ersonally found the ta lk in
te r e s tin g , highly in te re s tin g , or d u ll. W e want
your honest opinion.
No mention was made of the t e s t fo r inform ation th a t
was included on the b a llo t th a t was to follow the broadcast.
The preceding in s tru c tio n s were spliced to the tape th a t
held the appropriate introduction and the recorded ta lk fo r
use in the study.
The same s ta ff member who read the in s tru c tio n s
p rio r to the p rese n tatio n of the recorded m aterial to the
49
groups of su b je c ts a lso recorded the f i n a l in s tru c tio n s
which were sp liced to the tape a t the end of the rad io t a l k ,
following the passages of the song, "Easy to Love." The
follow ing announcement was used:
You w ill now be given your b a llo t. Please mark
your re a c tio n s to the t a l k as in te re s tin g , highly
in te r e s tin g , or d u ll. In a d d itio n , please mark your
answer to the tw enty-five questions on your b a llo t.
These questions are based upon the subject of;
sleep. From your own inform ation and opinion in ad
d itio n to the inform ation in the rad io t a l k , decide
on the answers th a t you believe are c o r r e c t. Please
do not put your name on the sheets given you. When
you have fin is h e d , give your paper to the student
a s s is ta n t in the stu d io . Thank you fo r coming to
day.
S ele ctio n of the su b je cts fo r
the experiment
The in v e s tig a to r decided to engage eig h ty su b je cts
in the main p a rt of t h i s study; one group of f o r ty su b je cts
was to have the speaker presented as the news commentator,
and the other group was to hear the same speaker as the
physician. Each subject in one group was paired d i r e c tly
with a subject in the second group on the b a s is of age, sex,
year in c o lle g e , previous te rm 's grade-point average, and
ACE placement score in the college entrance examina-
50
/ I
tio n s . Through cooperation with the ch ief co u n sello r of
each school on the campus of Oregon S tate College, the
health records of each student were examined fo r any gross
hearing d e fic ie n c ie s th a t were indicated in the stu d e n t's
record. The experim enter's colleagues in the Department of
Speech a s s is te d him s u b s ta n tia lly .
The subjects were drawn from the twenty sections of
e n ro lle e s in basic speech courses during the w inter term of
1958. In these c la sse s were 368 freshmen and sophomores.
The previous te rm 's GPA was obtained fo r each student who
had a fre e period on e i th e r Tuesday or Thursday at 10:00
A.M. The r e g i s t r a r 's office provided a l i s t of the ACE
placement scores of the students. There were 160 freshmen
and sophomores who indicated t h e i r a v a i l a b i l i t y fo r the ex
periment on one of the two morning hours th a t were required
fo r the te s tin g .
When a l l necessary inform ation on the 160 p o te n tia l
subjects had been gathered, twenty freshmen male subjects
were matched f o r age, year in co lleg e , ACE placement score,
and previous te rm 's GPA. Twenty freshmen female subjects
were matched with each other f o r the same c r i t e r i a . Then
twenty male sophomores were matched with each other and
twenty female sophomores were matched as c lo se ly as possible
^ American Council on Education Psychological Exam
in atio n fo r 'CoTl'e'qe'Ere shmen (l9b4 e d itio n ; Princeton:
Educational T esting Service, 1954).
fo r these c r i t e r i a . While a la rg e r sampling of the popula
tio n would be d e s ira b le , the selected subjects appeared to
be f a i r l y rep rese n tativ e of the general college population
of Oregon State C ollege. As reported in Tables 7 and 8,
the subjects were matched very c lo s e ly , there being a d if
ference of no more than .21 in previous term GPA fo r any
p a ir of su b je cts. Although prelim inary te s tin g had estab
lished a s ig n ific a n t re la tio n s h ip between GPA and scores on
the re te n tio n t e s t , t h i s in v e stig a to r f e l t th a t matching by
both previous-term GPA and ACE placement score would assure
a very close p airin g of the subjects fo r general i n t e l l i
gence. The se le c tio n of the subjects was accomplished dur
ing the f i r s t three weeks of the w inter term of 1958.
Physical se ttin g
The groups of subjects were scheduled to l i s t e n to
the rad io ta lk while seated in the radio studio in Shepard
H all. The studio appeared to be a more lo g ic a l se ttin g
than the ordinary classroom. The atmosphere of the studio
was pleasant and comfortable. The room was decorated in
subdued tones of green and brown and had a t tr a c tiv e p a in t
ings on the w alls. W all-to-w all green carpeting covered the
flo o r. The studio was soundproofed; i t had in d ire c t l ig h t
ing and e f f i c i e n t temperature c o n tro l. The fu rn itu re was
more appropriate in simulating a home-like atmosphere than
the c h a irs th a t were found in ty p ic a l classrooms. By using
52
TABLE 7
M ATCHED PAIRS OF M ALE SUBJECTS
USED IN THIS EXPERIMENT
A Aftr c - Previous . Previous
Age ACE Score Term Age ACE Score Term QpA
18 3 1.63 18 3 1.56
18 4 1.67 18 4 1.73
18 4 2.08 18 4 2.00
18 4 2.27 18 4 2.19
18 5 2.47 18 5 2.33
18 5 2.50 18 5 2.55
18 5 2.60 18 5 2.70
18 6 2.69 18 6 2.80
18 7 3.38 18 7 3.29
18 9 3.56 18 9 3.75
19 3 1.82
19 3 1.80
19 4 2.00 19 4. 2.15
19 4 2.20 19 4 2.31
19 4 2.39 19 4 2.50
19 5 2.53 19 5 2.43
19 7 2.95 19 7 2.93
19 7 3.12
19 7 3.10
19 7 3.29 19 7
3.22
19 7 3.33 19 7 3.30
19
8 2.80 19 8 3.00
NOTE: The three columns on the l e f t re p re se n t the
group of male su b jects who heard Chet P arker, the news com
mentator; the th ree columns on the r ig h t rep rese n t the group
of male su b je cts who heard Dr. Maloney, the physician.
53
TABLE 8
M ATCHED PAIRS OF FEMALE SUBJECTS
USED IN THIS EXPERIMENT
Age ACE Score
Previous
Term GAP
Age ACE Score
Previous
Term GAP
18 3 1.81 18 3 1.81
18 4 2.15 18 4 2.00
18 5 2.36 18 5 2.29
18 5 2.49 18 5 2.38
18 5 2.65 18 5 2.44
18 6 2.13 18 6 2.00
18 6 2.56 18 6 2.71
18 8 3.06 18 8 3.22
18 8 3.35 18 8 3.28
18 9 3.19 18 9 3.29
19 4 2.00 19 4 2.10
19 5 2.19 19 5 2.07
19 5 2.42
19 5 2.47
19 6 2.43 19 6 2.36
19 6 2.59 19 6 2.62
19 6 2.70 19 6 2.71
19 6 2.85 19 6 2.76
19 7 3.10 19 7 3.25
19 9 3.60 19 9 3.50
19 9 4.00 19 9 4.00
NOTE: The th ree columns on the l e f t re p re se n t the
group of female su b jects who heard Chet Parker, news com
mentator; the th ree columns on the r ig h t rep rese n t the group
of female su b jects who heard Dr. Maloney, the physician.
54
the one studio i t was possible to standardize the physical
se ttin g in which a l l groups were te s te d .
There were te c h n ic a l advantages in using the stu
dio. A Magnecorder, Model PT6-AH was in s ta lle d in the con
t r o l room. This recorder was connected d ir e c tly to a Gates
c o n tro l console, so th a t a tape-recording could be played on
the recorder with i t s volume co n tro lled through the Gates
console. The lis te n e r in the studio could not see the re
corder in the c o n tro l room. The studio audience received
the sound through a sen sitiv e Meissner speaker th a t was
b u ilt in to the studio w all. Technical q u a lity of the broad
c a s ts to the subjects could be r ig id ly c o n tro lle d .
Procedure s
Testing the groups
In order to r e ta in the te c h n ic a l q u ality of the
recordings m aterials used in the experiment, the introduc
tio n to the t a l k was re -s p lic e d to the recorded speech each
time th a t the schedule made i t necessary to change the in
tro d u ctio n . In order to minimize the necessity of re s p lic
ing the tape recording and the p o s s i b i l i t i e s of damaging the
m aterial on the ta p e, the groups of su b jects who were to
hear Chet Parker and Dr. Maloney were scheduled fo r a l t e r
nate weeks. One Tuesday morning th a t was not scheduled
o rig in a lly was necessary to complete the experiment because
five su b jects were absent at the periods when they should
have appeared. Table 9 in d ic a te s the schedule of the exper-
ime n t .
Time Period
TABLE 9
STUDIO SCHEDULE FOR THE 16 GROUPS
OF SUBJECTS
S u b jects Who Heard
Dr. Maloney
S ubjects Who Heard
Chet Parker
F i r s t Week
Second Week
Third Week
Fourth Week
F if t h Week
0
18
0
19
3
19
0
19
0
2
The eig h ty su b je c ts were c a lle d through the te a c h e rs
in whose basic speech sectio n s the su b je c ts were en ro lle d
and were n o tif ie d of a sp e c ific day and time f o r appearance
in the stu d io . Care was taken to schedule a l l students
drawn from a p a r tic u la r c la s s fo r the same day in order to
minimize passing of inform ation regarding the experiment
among the su b je cts. L ater a n a ly sis made of the scores on
the t e s t of inform ation revealed no s ig n ific a n t d iffe ren ce
favoring the su b je cts who were scheduled during the t h ir d
and fo u rth weeks of the experiment from those who p a r t i c i
pated during the f i r s t and second weeks. If th ere was any
passing of inform ation as the experiment progressed, i t was
not in d icated by the t e s t scores. The comparative data upon
which t h i s conclusion was based are reported in the Appen
dix.
The su b jects were to ld th a t they would p a rtic ip a te
titrv - •
in an experiment requiring no prep aratio n by them. They
were also to ld th a t i t was important th a t they appear a t the
studio on time. No fu rth e r in s tru c tio n was given. In order
to minimize the possible e f f e c ts of classroom -like group
lis te n in g , the size of each group was lim ited to fiv e sub
je c ts . In every group both sexes were represented in order
to simulate as r e a l i s t i c a l l y as p ra ctic a b le a group of per
sons who might be found in a family se ttin g . In some groups
were two subjects of one sex and three of the opposite sex;
in other groups were four lis te n e r s of one sex and one of
the opposite sex.
When the group of subjects appeared at the studio a t
the scheduled tim e, a student a s s is ta n t took the group in to
the studio where the su b jects were to ld to make themselves
com fortable. Five c h a irs were placed in the studio a t
specific d istan ce s from the wall speaker, assuring a l l sub
je c ts of good audio recep tio n . There were no in d ic atio n s
during the conduct of the experiment th a t seating arrange
ments had any e f f e c t on the r e s u lt s of the study. As a l a t
e r check on the a c tu a l e f f e c ts of seating arrangement,
scores and le v el of in te r e s t of those subjects who were
seated in c h a irs 2, 3, and 5 as indicated in Figure 1 were
recorded and compared. R esults were almost id e n tic a l fo r
57
C ontrol Room Window
Speaker in Wall
( ) A s s is ta n t’s c h a ir
5
2
Fig. 1 .— Studio seating arrangement f o r
su b jects and a s s is ta n t with provision fo r
recording names of su b jects by a s s is ta n t.
b&
the three groups with no group showing a sig n ific a n t d i f f e r
ence in t e s t scores or le v el of in te r e s t. A ta b le of t h i s
data w ill be found in the Appendix. When the group of sub
j e c t s had been seated, the a s s is ta n t then took r o l l , marking
on her paper the order in which each p a r tic ip a n t had seated
him self. This inform ation was recorded as unobtrusively as
possible on a seating chart sim ilar to the one shown in Fig
ure 1.
No paper or te s tin g m aterial was v is ib le in the
studio, with the exception of the slip of paper on which the
a s s is ta n t had taken r o l l . The student a s s is ta n t then sat in
a c h a ir in one corner of the studio and made h e rse lf as in
conspicuous as p o ssib le.
The ta p e-re co rd er in the co n tro l room was turned on,
and the su b jects heard the broadcast through the wall
speaker in the radio studio. The recordings included the
opening explanation, the in tro d u ctio n of the speaker, the
radio ta lk , and the concluding in s tru c tio n s . The running
time of the recording as Chet Parker amounted to nine min
u te s and t h i r t y seconds; the running time of the tape in
which the in tro d u ctio n of Dr. Maloney was su b stitu te d fo r
th a t of Chet Parker was nine minutes and twenty-seven sec
onds.
A fter the concluding in s tru c tio n s on the ta p e -re
cording were played, the a s s is ta n t in the studio d is trib u te d
the t e s t m a terials and returned to her seat in the studio.
59
As each subject completed the answers to the q u estio n s, the
studio a s s i s t a n t took the s u b je c t's paper, making c e r ta in
th a t the paper and the s u b je c t's name on the seating a r
rangement corresponded. This procedure was observed by the
in v e s tig a to r from the c o n tro l room so th a t e r r o r would be
avoided in id e n tify in g the examination paper with the proper
su b je ct. Without being under pressure of any kind, a l l sub
j e c t s completed the examination in le s s than twelve minutes.
In ad d itio n to the in v e s tig a to r who was in the con
t r o l room and the a s s is ta n t who was in the studio* two other
students a s s is te d with the study. One student served in the
c o n tro l room as an a s s is ta n t engineer; the other student re
mained in the hallway outside of the studio to prevent any
in te rr u p tio n th a t might in te rf e r e with the experiment and
also to co n tact the students as they a rriv e d to p a r tic ip a te
in the experiment. Between 10:25 and 10:30 on the scheduled
days in te rru p tio n was very lik e ly to occur, if not fo r the
help of the a s s is ta n t who was in the h a l l . By using the
same personnel during the fiv e weeks of the experim ent, no
serious problem was encountered.
S t a t i s t i c a l methods
T his in v e s tig a to r had two devices of measurement
av ailab le f o r determining the e f f e c t s of the radio t a l k upon
the su b je c ts: (1) the r a tin g scale of in te r e s t in the in
form ation and (2) the scores made by the su b jects on the
60
r e te n tio n t e s t . On the b asis of inform ation obtained from
these two measurement devices, seven n u ll hypotheses were
te s te d in the f i n a l p a rt of t h i s study. The basic nu ll
hypothesis a s s e r ts th a t there is no tru e d ifferen ce between
two means, and th a t the d ifferen ce found between sample
means is th e re fo re ac cid e n tal and unimportant. If the
hypothesis is untenable, i t must be re je c te d .
For determining the le v e l of in te r e s t in the t a l k ,
the c h i square t e s t represented a u se fu l method of comparing
experim entally obtained r e s u l t s with those to be expected
th e o r e tic a lly on the b a sis of a n u ll hypothesis. The d i f
ference between observed and expected frequencies are
squared and divided by the expected number in each case, and
o
the sum of these q u o tien ts is X . The more c lo se ly the ob
served r e s u lt s approximate to the expected the smaller the
chi square and the c lo s e r the agreement between observed
data and the hypothesis being te s te d . The la rg e r the chi
square, the g re a te r the p ro b a b ility of a r e a l divergence of
experim entally observed from expected r e s u l t s . This s t a t i s
t i c a l method was appropriate because the observations of
in te r e s t were dicho'tomous in stead of q u a n tita tiv e .
A c o r r e la tio n e x is ts between two samples of scores
when they represent the performance of two matched groups,
7
Henry E. G a rre tt, S t a t i s t i c s in Psychology and
Education (5th ed .; New York: Longmans, Green and Company,
1958J, p. 213.
an experim ental and a c o n tro l group. This in v e stig a to r used
the p aire d t - t e s t to compare r e te n tio n scores with each
Q
other. This procedure was recommended by G uilford if the
subjects of comparison are clo sely matched by p a ir; the
chairman of the Department of S t a t i s t i c s a t Oregon State
College concurred in the recommendation. The desired sta
t i s t i c s were obtained d ir e c tly from d iffe re n c e s between
p a irs . If we sum the d iffe re n c e s and divide by N, we obtain
the mean of the d iffe re n c e , which is equal to the d ifference
between the means. The SE of the mean of these d iffe re n c e s
was obtained d ir e c tly from the d iffe re n c e s. This method is
explained by G uilford as the most d ir e c t where computation
machines are a v a ila b le . The paired t - t e s t was used in t e s t
ing hypotheses. The data were corrected fo r e rro rs of
omission by use of a standard formula fo r t h i s purpose.9
The general objective of an a ly sis of variance pro
cedures is to t e s t fo r s ig n ific a n t d iffe re n c e s between sev
e r a l se ts of independently derived experim ental samples to
see whether they could or could not have a ris e n by random
sampling from the same population. Two d is t i n c t estim ates
of the population variance are made, one derived from w ithin
the sum of squares and the other from the "between" sum of
Q
J. P. G uilford, Fundamental S t a t i s t i c s in Psychol
ogy and Education (2nd ed.; "N ew York: McGraw-Hill Book
Company, In c., 1950), pp. 219-20.
9Ib id . . pp. 533-34.
squares. If these two estim ates are very sim ilar, one is
in clin ed to accept the n u ll hypothesis th a t the two s e ts of
measurements did a rise from the same population. If the two
estim ates d i f f e r s u f fic ie n tly , the hypothesis is re je c te d .
The a n a ly sis of variance procedure was used where the data
involved two se ts of f a c to rs fo r which t h i s procedure was
the proper s t a t i s t i c a l method.
Summary
The purpose of t h i s study required th a t the experi
mental method of research be used. Health and medical as
pects of the topic of sleep were selected fo r development
into a radio ta l k . The ta lk was delivered by a reader who
was selected on the b a s is of p re -te s tin g procedures which
indicated th a t he would be acceptable to the l is te n e r s i f he
were introduced as a radio newscaster or as a physician.
Two instrum ents were devised and te s te d fo r purposes of
measuring the amount of understanding reported by the sub
j e c t s and the lev el of in te r e s t in the inform ation.
Eighty subjects were matched fo r age, sex, year in
co lleg e , previous te rm 's grade-point average, and ACE place
ment score. Forty su b jects heard the speaker introduced as
the physician and fo rty subjects heard the speaker in tro
duced as the newscaster. The in tro d u ctio n of the speaker
was the only manipulated variable in the experiment. Sub
j e c ts in groups of fiv e heard the radio ta lk in the radio
studio under c lo sely co n tro lled conditions. Following the
p re se n ta tio n of the ta lk , a l l subjects were given the t e s t
fo r re te n tio n of inform ation and the th ree -p o in t ra tin g
scale of in te r e s t.
Seven n u ll hypotheses were te s te d in the f i n a l phase
of the study. For determining the le v el of in te re s t in the
ta lk , the chi-square t e s t of independence was used. This
in v e s tig a to r employed the paired t - t e s t to compare re te n tio n
scores with each other. The an aly sis of variance procedure
was used where the data involved two se ts of fa c to rs for
which t h i s procedure was the proper s t a t i s t i c a l method.
CHAPTER I I I
PRESENTATION AND INTERPRETATION OF DATA
Raw Data
Eighty students were matched fo r age, year in c o l
lege, sex, grade-point average of previous term and ACE
1
placement score. These su b je cts were drawn from the en-
r o lle e s in the basic speech courses a t Oregon S ta te C ollege.
None of these students had p a r tic ip a te d in any of the pre
lim inary experiments involved in t h i s study. Forty of these
su b je cts were males and f o r ty were fem ales. Of the eig h ty
su b je c ts , f o rty were freshmen and f o r ty were sophomores.
One group of f o r ty su b jects was exposed to the rad io t a l k
th a t was introduced as a p re se n ta tio n of Dr. Ross Maloney, a
physician in research a t the Denver Municipal H o sp ital, The
other group of fo r ty su b je cts heard the rad io t a l k in tr o
duced as a rep o rt by Chet P arker, lo c a l news commentator and
an a ly st fo r a Denver ra d io s ta tio n . The only manipulated
v ariab le in the experim ental design was the id e n tif ic a tio n
of the speaker.
*See Appendix f o r ta b le of paired d a ta , including
sex, year in c o lle g e , age, grade-point average in the p re v i
ous term , ACE placement score, degree of in te r e s t in the
p re s e n ta tio n , and raw score on the t e s t of inform ation.
64
65
In order to evaluate the responses of the su b jects
to the p re se n ta tio n of the radio t a l k , two p o s t- te s t or
2
" a fte r-o n ly ” instrum ents were used. Each subject was asked
to check a th re e -p o in t ra tin g scale of h is in te r e s t in the
radio ta lk ; and each subject was given a re te n tio n t e s t of
the inform ation in the t a l k . Copies of the te s tin g mater
i a l s w ill be found in the Appendix. For purposes of con
venience the group who heard Dr. Maloney w ill be termed the
control-group in the in te rp r e ta tio n of the data.
Testing the F i r s t Null Hypothesis
The f i r s t n u ll hypothesis was sta ted as follows:
As measured by a re te n tio n t e s t , understanding
of the radio ta l k on sleep fo r a c o n tro l group of
su b jects who heard a physician did not d if f e r sig
n if ic a n tly from th a t of those subjects who lis te n e d
to a newscaster.
The paired t - t e s t indicated th a t there was a true
difference between the le v e ls of understanding in the two
groups of su b je cts. As reported in Table 10, the observed
value of t is 2.113 with 39 d .f . Although the difference
in the scores on the re te n tio n t e s t were not s ig n ific a n t at
the 1 per cent le v e l, they were sig n ific a n t at the 5 per
^Carl I. Hovland, Arthur Lumsdaine, and Fred Shef
f i e l d , Experiments on Mass Communication. Vol. I l l : S tu d ies
in S ocial Psychology in World War I I f4~vols.; Princeton:
Princeton U niversity P ress, 1949-50), p . 28.
66
TABLE 10
PAIRED t ANALYSIS OF SCORES* ON TEST
OF UNDERSTANDING
Dr. Maloney Chet Parker d
21 20 1
22 21 1
19
22 -3
20
21 -1
22 22 0
22 22 0
21 22 -1
23 21 2 d = .59125
22 22 0
24 22 2 s = 1.7695
20.33
19
20
21
.33
-2
t = *5 9125 _ 2.113
25 21 4
1.7695
23 19 4
C r i t i c a l value of t
20.33 22 -1 .6 7
22 22 0
with 39 d . f . = * 2.023
22 21 1
23
23.33
23
22
0
1.33
a t the 5# le v e l of
22
23
23
21
-1
2
conf idence.
22 22 0
21 22 -1
23 19 4
25 21 4
22.67 22 .67
21.33 20 1.33
24 24 0
23 24 -1
23
22 1
21 18.67 2.33
22 21 1
23 20 3
20 20 0
21.33 23 -1 .6 7
22 23 -1
22 21 1
23 20 3
21 22 -1
21 22
23.6b'
*Scores were c o rre c te d f o r e r r o r s of omission using
67
cent le v e l. Because the value of t is la rg e r than the
c r i t i c a l value of 2.023, the hypothesis was re je c te d . I t
may be concluded th a t the group of su b jects who lis te n e d to
Dr. Maloney had g re a te r understanding of the ta l k than the
group of su b jects who heard the newscaster, Chet Parker.
T esting the Second Null Hypothesis
What re la tio n s h ip , if any, ex iste d between the un
derstanding of the inform ation and the le v el of in te r e s t
3
th a t was reported by the subjects? In h is study of author-
ity - p re s tig e th a t used a h alf-h o u r rad io program which in
corporated two rad io t a l k s , Highlander found no difference
between h is co n tro l group and h is experim ental group in re
te n tio n of inform ation. On the b a sis of h is fin d in g s in us
ing an e le c tro n ic program analyzer, he suggested th a t there
appeared to be a re la tio n s h ip between audience in te r e s t in
the subject m atter and the a b i l i t y to r e c a l l inform ation.^
On the other hand, in reporting on the B ritis h
formula suggested by J . P. G uilford; exact reference is
given in Chapter I I . The lack of balance on the number of
item omissions (ten fo r su b jects who lis te n e d to the physi
cian and two fo r subjects who heard the newscaster) j u s t i
fie d the use of a c o rre c tio n formula. The need fo r t h i s
c o rre c tio n formula did not a ris e in the other comparisons
made in the study.
3
Supra, p. 33.
4John P. Highlander, "Audience Analyzer Measure
ments and Inform ational E ffec ts of Speaker V ariables in
Radio T alks," Speech Monographs. XXI (August, 1954), IBS-
89.
68
stu d ie s of rad io t a l k s , Trenaman found no s ig n ific a n t r e l a
tio n s h ip between scores of r e c a l l of inform ation and in te r -
e s t r a tin g s . Was th ere any s ig n ific a n t r e la tio n s h ip be
tween t e s t scores and in t e r e s t le v e ls of the two groups of
su b je c ts in t h i s experiment? The follow ing nu ll hypothesis
was te s te d :
The le v el of in te r e s t rep o rted by those su b jects
who lis te n e d to the ta l k d eliv ered by the subject-
m atter s p e c ia lis t did not d i f f e r s ig n if ic a n tly from
the le v e l of in t e r e s t in the inform ation rep o rted
by the su b je cts who heard the newsman.
A ll su b je cts marked the b a llo t fo r degree of in te r-
e s t , and no subject marked the t a l k as d u ll. As shown in
Table 11, 19 members of the c o n tro l group found the ta lk
in te r e s ti n g , while 21 rep o rted the t a l k as highly i n t e r e s t
ing. In the experim ental group, 29 found the ta l k i n t e r e s t
ing and only 11 marked the t a l k as hig h ly in te r e s tin g .
The chi-square t e s t of independence was used to
measure the d iffe re n ce in le v e l of i n t e r e s t reported by the
c o n tro l and experim ental groups of s u b je c ts . The computed
chi-square is 5.21 with one degree of freedom. Because
t h i s value is g re a te r than 3.84, which in d ic a te s s i g n i f i
cance a t the 5 per cent le v e l, the hypothesis was re je c te d .
^Joseph Trenaman, "Understanding Radio T alk s," The
Q u arterly Jo u rn al of Speech. XXXVII (A pril, 1951), 178.
^Supra. p. 33.
69
TABLE X X
CHI-SQUARE TEST OF INTEREST LEVELS
OF CONTROL AND EXPERIMENTAL GROUPS
W H O HEARD THE RADIO TALK
Speaker Intere sting Highly In te re s tin g T otaX
Dr. Maioney
(A)
X9
(B)
2X
(A
+ B)
40
Chet Parker
(C)
29 '
■(D)
X X
(c
+ D)
40
(A + C) (B + D) N
48 32 80
X
n(ad- bc)
2
(A+B) (C+D) (A+C) (B+D)
= 5.2X with X d .f
2
C r i t i c a l vaiue of X >3.84 fo r b % ie v ei of sig n ific an c e.
70
I t may be concluded th a t the average le v e l of in t e r e s t of
the su b jects who heard Dr. Maloney was higher than the aver
age le v e l of in te r e s t th a t was in d ic ated by the su b je c ts who
lis te n e d to the ta l k as a t tr i b u te d to Chet Parker, the news
commentator.
T esting the Third Null Hypothesis
Those su b je cts who heard Dr. Maloney rep o rted both
higher scores on the r e te n tio n t e s t and g re a te r average in
t e r e s t as measured by the th re e -p o in t r a tin g sc ale. Was the
d ifferen ce in the inform ation scores r e la te d to a d iffe re n ce
in le v e l of in te r e s t in the topic of the t a l k , or to a d i f
ference in resp e ct f o r the id e n tity of the speaker and
th e re fo re a d iffe re n ce in a tte n tiv e n e s s to the inform ation?
Although i t was highly improbable, th e re was a s lig h t pos
s i b i l i t y th a t the su b je cts who were selec ted f o r the c o n tro l
group had a much g re a te r in t e r e s t in the to p ic than the sub
j e c t s who were chosen f o r the experim ental group. T his in
v e s tig a to r did not give the se lected su b je cts an opportunity
to r e g i s t e r in advance of the t e s t t h e i r degree of i n t e r e s t
in the to p ic of sleep. I t was believed th a t a p r e - te s t of
i n t e r e s t in the topic of sleep would not measure i n t e r e s t in
the t a l k i t s e l f and might influence subsequent ev a lu a tio n of
the to p ic as presented in the form of a recorded rad io pro
gram. From the p relim inary e v a lu a tio n s rep o rted in Chapter
I I , t h i s in v e s tig a to r was assured with reasonable c e r t a in ty
71
th a t co lleg e students would have some in t e r e s t in the topic
of sleep. I t would be d o u b tfu l, a ls o , th a t one of the
groups of matched su b je c ts had a c o n s is te n tly stronger pre
d is p o s itio n to be in te re s te d in the topic of sleep than the
other group.
In an attempt to separate i n t e r e s t in the to p ic from
the id e n tity of the speaker, the follow ing n u ll hypothesis
was te s te d :
Among matched p a irs of su b je c ts who marked the
t a l k as " in te re s tin g " r a th e r than "highly i n t e r e s t
ing," understanding of the ra d io t a l k on sleep fo r
the c o n tro l group of su b je cts who heard the sp e c ia l
i s t did not d i f f e r s ig n if ic a n tly from th a t of the
su b je cts who heard the newscaster.
If the scores r e g is te r e d no s ig n ific a n t d iffe re n ce
in understanding between the two groups, t h i s would suggest,
w ithin the lim ita tio n s of the small sample of the t h i r t e e n
matched p a irs th a t could be stu d ied , th a t the topic instead
of the speaker might be the determ iner of the d ifferen ce in
inform ation th a t appeared between the two groups of sub
j e c t s .
The paired t - t e s t was employed as the proper s t a t i s
t i c a l method. As shown in Table 12, the c r i t i c a l value of
t was 2.179. The observed value of t was 2.28 with twelve
degrees of freedom. This value was s ig n ific a n t a t the 5 per
cent le v e l of confidence. The hypothesis th e re fo re was re-
72
TABLE 12
PAIRED t ANALYSIS OF SCORES ON TEST
OF UNDERSTANDING OF THOSE M ATCHED PAIRS
OF SUBJECTS W H O M ARKED THE RADIO TALK
AS INTERESTING
Subjects Who Heard
Dr. Maloney
Subjects Who Heard
Chet Parker
d
23 21 2
23 19 4
23 20 3
22 21 1
22 22
0
22 22 0
22 22 0
22 21 1
21 20 1
21 22 -1
21 18 3
21 22
-1
20 20 __0
13
N = 13 s /N = .1923
(2y)2N = 13.00 V s^N = .43852
Jy2 = 4 3 7 = 1.0000
SS = 30.00 t = 2.28 with 12 d .f .
s2 = 2.50 C r i t i c a l value of t is
2.179 at 5$ le v el of
conf idence
73
je c te d . I t then may be concluded th a t of those matched
p a irs who marked the ta lk as in te re s tin g , subjects who were
in the c o n tro l group made s ig n ific a n tly higher scores on the
re te n tio n t e s t than those subjects who heard Chet Parker.
The id e n tity of the speaker and not the in te r e s t in the
topic re s u lte d in g re a te r understanding of the inform ation
among the c o n tro l group which heard Dr. Maloney.
T esting the Fourth Null Hypothesis
One other avenue was explored by t h i s in v e stig a to r
in order to determine whether the higher scores on the re
te n tio n t e s t were asso ciated with in te r e s t in the topic in
stead of the id e n tity of the speaker. Of the eig h ty sub
j e c t s in the experiment, twenty-two who reported the ta l k
as highly in te re s tin g were pair-m ates of su b jects who in
dicated th a t the ta l k was in te re s tin g . If degree of in te r
e s t in the topic of the ta l k instead of the id e n tity of the
speaker determined the understanding of the inform ation,
then those su b jects who found the ta lk highly in te re s tin g
should have made s ig n ific a n tly higher average scores on the
t e s t of inform ation than did t h e i r matched counterparts who
found the t a l k in te re s tin g . The following hypothesis was
te s te d :
The average scores on the re te n tio n t e s t fo r the
subjects who marked the ta l k as "highly in te re s tin g "
did not d i f f e r s ig n ific a n tly from the scores made by
t h e i r matched c o u n te rp a rts who marked the t a l k as
"in tere s t i n g . "
The p aire d t - t e s t was used to t e s t t h i s hypothesis.
As in d icated in Table 13, the observed value of t was 1.55
with 21 d . f . Since the c r i t i c a l value of t was 2.080 with
21 d . f . , the observed value of t was in s u f f ic ie n t to be sig
n if ic a n t a t the 5 per cent le v e l. Therefore the hypothesis
was accepted. The su b je c ts in the two groups who marked the
ta l k as highly in te r e s tin g did not make b e t t e r average
scores on the r e te n tio n t e s t than t h e i r matched counter
p a rts who marked the ta lk as in te r e s ti n g . This r e s u l t would
appear to support the conclusion th a t fo r these fo rty - f o u r
su b jects the id e n tity of the speaker in ste a d of in t e r e s t in
the to p ic determined the average amounts of inform ation th a t
were re ta in e d . The i d e n tity of Dr. Maloney re s u lte d in more
understanding of the inform ation and g r e a te r in t e r e s t in the
ta l k than did the id e n tity of Chet P arker. There is no e v i
dence to support the assumption th a t the in t e r e s t le v el
among the c o n tro l group was higher than th a t among the ex
perim ental group because the two groups were d i f f e r e n t l y pre
disposed to be in te re s te d in the topic of sleep.
Testing the F if t h Null Hypothesis
In previous stu d ie s of e f f e c t s of a u th o rity -p re s tig e
upon s h if t of opinion and r e c a l l of inform ation, l i t t l e con
sid e ra tio n was given to d iffe re n c e s by sex in the su b je ct.
75’
TABLE 13
PAIRED t ANALYSIS OF TEST SCORES
OF THOSE M ATCHED PAIRS OF SUBJECTS
W HERE ONE SUBJECT FOUND THE TALK
HIGHLY INTERESTING
AND HIS M ATCHED COUNTERPART FOUND
THE TALK INTERESTING
Scores of Those Who Scores of Those Who
Found Talk Highly Found Talk d
I nte re st ing Inte re st ing
25 21 4
24 24 0
24 2 2 2
23 23 0
23 19 4
23 2 2 1
23 2 2
1
23 20 3
23 24 -1
22 22
0
22 22
0
22 20 2
22 21 1
22 21 1
22 23 -1
21 19
2
21 23 - 2
21 20 1
21 22 -1
20
21 -1
20 20 0
19
22
zl
13
CM
CM
1 1
~ \ l s 2 /N = .3 8 1 7
( £ y ) 2 N = 7 .6 8
y = .5 9 0 9
£ y 2 = 75 t = 1 .5 5 with 2 1 d .f .
SS - 6 7 . 3 2
o
C r i t i c a l value of t is
2 .0 8 0 a t 5 % le v e l of
s = 3 .2 0 5 7 confide nee.
s2/N = . 1457
76
This in v e s tig a to r studied equal numbers of male and female
su b je cts in order to make the s e ttin g of the lis te n in g like
th a t of lis te n in g in fam ily groups. On prelim inary te s tin g
of the rad io t a l k , the topic appeared to be of general
i n t e r e s t ; but the v aria b le of the id e n tity of the speaker
was not introduced in to the prelim inary e v a lu a tio n s. In or
der to discover the ex ten t to which d iffe re n ce in sex was
asso ciated with le v e l of in te r e s t in the inform ation a f t e r
the in tro d u c tio n of the v aria b le of s p e a k e r-id e n tity , the
follow ing nu ll hypothesis was te s te d :
The le v e l of in te r e s t in the inform ation re
ported by the group of male su b je cts did not d i f f e r
s ig n if ic a n tly from the average le v e l of in te r e s t
reported by the group of female su b je c ts .
A chi-square t e s t of independence was used. As re
ported in Table 14, the c r i t i c a l value of was 3.84. The
computed chi-square value was .83 with 1 d . f . Because t h i s
fig u re i s not s ig n ific a n t a t the b per cent le v e l, the
hypothesis was accepted. I t may be concluded th a t the aver
age le v e l of in te r e s t of the male su b je cts in t h i s ex p e ri
ment did not d i f f e r s ig n if ic a n tly from the average le v e l of
in te r e s t in the ta l k rep o rted by the female su b je cts. Al
though the group of su b jects who heard Dr. Maloney had sig
n if ic a n tly g re a te r in te r e s t in the inform ation than those
who heard Chet P arker, in te r e s t in the topic was asso ciated
in s ig n if ic a n tly with sex of su b je c t. S ele ctio n of the topic
11
TABLE 14
COMPARATIVE INTEREST LEVELS OF M ALE
AND FEMALE SUBJECTS USED IN THIS EXPERIMENT
Sex Intere sting Highly In te re stin g
T otal
Male s
(A)
26
(B)
14
(A + B)
40
Female s
(c)
22
(D)
18
(C + D)
40
(A + C)
48
(B + D)
32
N
80
N(AD-BC)
= .83 with 1 d .f .
(A+B) (C+Dj (A+C) (B+D)
2
C r i t i c a l value of X >3.84 fo r 5 % lev el of significance
78
of sleep appeared to be a sound choice in meeting the c r i
te rio n of general in te r e s t in a topic by both male and fe
male su b jects.
Testing the S ixth Null Hypothesis
Because the su b jects had been matched by both age
and year in college as w ell as other c r i t e r i a , the in te r e s t-
le v e l of the freshmen was compared to the i n t e r e s t - l e v e l of
the sophomores. There were th ir te e n freshmen who had marked
the t a l k as highly in te re s tin g to compare with nineteen
sophomores who had found the ta l k highly in te re s tin g . Of
the two age groups, twenty-seven freshmen and twenty-one
sophomores had indicated th a t the t a l k was in te re s tin g . The
following n u ll hypothesis was te s te d :
The average le v e l of in te r e s t in the radio ta l k
indicated by the freshman su b jects did not d i f f e r
s ig n ific a n tly from th a t of the sophomore su b je cts.
As shown in Table 15, the chi-square t e s t of inde
pendence was used. The c r i t i c a l value of chi-square was
3.84 a t the 5 per cent le v e l of confidence. The observed
value of chi-square was 1.875 with 1 d .f . Because t h i s
value was not sig n ific a n t at the 5 per cent le v e l, the hy
p o t h e s is was accepted. One may conclude th a t the average
le v el of in te r e s t indicated by the freshmen subjects did not
d i f f e r s ig n ific a n tly from the average lev el of in te r e s t in
the rad io t a l k as reported by the sophomore su b je cts.
79
TABLE 15
COMPARATIVE LEVELS OF INTEREST IN THE TALK
FOR FRESHMEN AND SOPHOM ORE SUBJECTS
USED IN THIS EXPERIMENT
College
C lass
In te re stin g
Highly
Intere sting
T o ta l
Fre shmen
(A)
27
(B)
13
(A + B)
40
Sophomore s
(c)
21
(D)
19
(C + D)
40
(A + C) (B + D) N .
48 32 80
i
X2 =
N(AD-BC)
2
(A+B) (c+d ) (a+c ) (b+d )
= 1.875 with 1 d .f .
C r i t i c a l value of X2 >3.84 fo r le v el of sig n ific an c e.
Testing the Seventh Null Hypothesis
The co n tro l group and the experim ental group had
been matched a lso fo r previous term grade-point average and
ACE placement scores. If these c r i t e r i a f o r matching the
su b jects properly indicated general in te llig e n c e , then the
subjects who earned higher ra tin g on the ACE t e s t lo g ic a lly
would make b e tte r scores in the r e c a l l of inform ation, de
spite the variable of the id e n tity of the speaker; i f t h i s
were not tr u e , e ith e r the c r i t e r i a th a t were used in match
ing in te llig e n c e or the v a l id ity of the t e s t of inform ation
would be suspect. To e s ta b lis h re la tio n s h ip between the
matching procedures f o r general in te llig e n c e and scores on
the t e s t of inform ation, the following n u ll hypothesis was
te s te d :
Among su b jects who were matched by ACE placement
scores w ithin the same te n th of the d is tr ib u tio n of
scores on the ACE t e s t of sch o lastic aptitude in
Oregon S tate C ollege, the average scores on the t e s t
of understanding by those whose AGE scores were be
low the sixth te n th did not d i f f e r s ig n ific a n tly
from those whose ACE scores were in or above the
six th te n th .
i
A nalysis of variance was used to determine the ac
ceptance or r e je c tio n of t h i s hypothesis. This set of data
shows a comparison of two c r i t e r i a : the ACE placement
scores, high and low, are one c r ite r i o n ; the two speakers, j
Parker and Maloney, are the o th e r.^ The a n a ly sis of v a ri- !
ance of the t e s t scores i s shown in Table 16. The r e s u l t !
shows th a t su b je cts who earned a high ACE ra tin g had s ig n if
ic a n tly higher scores in the t e s t of inform ation than did
those who earned a low ACE r a tin g . This d ifferen ce i s sig
n if ic a n t a t the 1 per cent le v e l. Therefore the hypothesis
is r e je c te d and one may conclude th a t the more i n t e l l i g e n t
|
as measured by ACE placement scores re ta in e d more informa
tio n than the le s s i n t e l l i g e n t su b je cts. In a v a lid t e s t of;
inform ation t h i s conclusion was to be expected.
Summary
Eighty students were matched fo r age, year in co l- ;
lege, sex, previous te rm 's grade-point average, and ACE
placement score. One group of f o rty su b je c ts was exposed i
to the ra d io t a l k th a t was introduced as a p re se n ta tio n of
Dr. Ross Maloney, a physician in research a t the Denver
i
Municipal H o sp ital. The other group of f o r ty su b je cts heard!
the rad io t a l k introduced as a rep o rt by Chet Parker, lo c a l
news commentator and an a ly st fo r a Denver rad io s ta tio n .
The only manipulated v a ria b le in the experim ental design was
' i
[the i d e n tif ic a tio n of the speaker. j
I In order to evaluate the responses of the su b je cts ;
i j
to the p re s e n ta tio n of the radio t a l k , two p o s t- te s t in s tru
ments were used. Each subject was asked to check a th re e -
7
See ta b le of raw data in Appendix.
82
i
TABLE 16 !
COMPARISON OF TEST SCORES OF SUBJECTS
W HOSE ACE SCORES W ERE IN THE SIXTH TENTH
OR ABOVE
WITH SUBJECTS W HOSE ACE SCORES W ERE
BELOW THE SIXTH TENTH
I
Source of
V ariatio n
Sum of
Squares
Degrees of
Freedom
Mean
Square
F ;
Tenth 12.800 1 12.800
!
8.98*
Within Tenth 54.150 38 1.425
i
Speaker 3.200 1 3.200 1.09
Tenth x
Speaker
.050 1 .050
.02
Error 111.100 38 2.924
t
T o ta l 139.950 79
!
i * S ig n ific a n t a t 1 % le v e l. and 1/6 F values with
1 and 38 d . f . are 4.09 and 7.32 re s p e c tiv e ly .
point ra tin g scale of h is in te re s t in the radio ta lk ; and
each subject was given a re te n tio n t e s t of the inform ation
in the t a l k .
84'
TABLE 17
RESULTS OF TESTING SEVEN NULL HYPOTHESES
„ ,, . Test of
Hypothesis Significance*
As measured by a re te n tio n t e s t , understand
ing of a rad io ta lk on sleep fo r the c o n tro l
group of subjects who heard a physician did s
not d if f e r s ig n ific a n tly from th a t of those
subjects who liste n e d to a newscaster
The lev el of in te r e s t reported by those sub
j e c t s who lis te n e d to the ta l k delivered by
the sub ject-m atter s p e c ia lis t did not d if f e r
s ig n ific a n tly from the le v el of in te r e s t in
the inform ation reported by the subjects who
heard the newsman.
Among the matched p a irs of subjects who
marked the ta lk as " in te re s tin g " r a th e r than
"highly in te re s tin g ," understanding of the
radio ta l k on sleep fo r the co n tro l group of s
subjects who heard the s p e c ia lis t did not
d if f e r s ig n ific a n tly from th a t of the sub
je c ts who heard the newscaster.
The average scores on the re te n tio n t e s t fo r
the subjects who marked the ta l k as "highly
in te re s tin g " did not d i f f e r s ig n ific a n tly
from the scores made by t h e i r matched coun- ns
te r p a r ts who marked the ta lk as " in te r e s t
ing ."
The lev el of in te re s t in the inform ation re
ported by the group of male subjects did not
d i f f e r s ig n ific a n tly from the average le v e l ns
of in te re s t reported by the group of female
subjects.
The average le v el of in te re s t in the radio
ta lk indicated by the freshman subjects did
not d i f f e r s ig n ific a n tly from th a t of the ns
sophomore subjects.
85
TABLE 17—C o n tin u ed
Hypothe s is
T e st of
S ig n if icance*
Among su b je cts who were matched by ACE
placement scores w ithin the same te n th of
the d is t r ib u t io n of scores on the ACE t e s t
of sc h o la stic aptitude in Oregon S ta te
C ollege, the average scores on the t e s t of
understanding by those whose ACE scores were
below the six th te n th did not d i f f e r sig
n if ic a n tly from those whose ACE scores were
in or above the sixth te n th .
vs
*vs = very s ig n if ic a n t (.01 le v el)
s = s ig n ific a n t (.05 lev el)
ns = not s ig n ific a n t
CHAPTER IV
SUM M ARY, CONCLUSIONS, IMPLICATIONS,
AND RECOMMENDATIONS
Summary
Despite the sig n ific an c e of the broadcast media in
the United S ta te s , most of our knowledge of the e f f e c t s of
rad io in the d issem ination of inform ation has been based
upon nonexperimental s tu d ie s . The l i t e r a t u r e re v e a ls l i t t l e
research a n a ly sis of the v a ria b le s th a t make one informa
tio n a l program more e ffe c tiv e than another. As indicated by
Hovland, Merton, and o th e rs, there is a s u b s ta n tia l lack of
knowledge of the e f f e c t s of the ra d io communicator upon the
l i s t e n e r .
P ro fessio n al broadcasters and educators have d i f
fered in th e i r views of the importance of the speaker in
the p re s e n ta tio n of the inform ative ra d io t a l k . Some have
contended th a t the i n t e r e s t of the li s t e n e r in the content
is most im portant, b eliev in g th a t i f th e re is in te r e s t in
the to p ic and the m a teria l is well prepared, the voice of
the p ro fe ssio n a l ra d io announcer w ill impart as much know
ledge as the voice of the s p e c i a l i s t . O thers have in s is te d
th a t the voice of the in fo rm a tio n -s p e c ia lis t adds p re stig e
86
87
to the topic and produces more a tte n tio n and g re a te r under
standing than the voice of the announcer.
I t was the purpose of th i s study to provide informa
tio n with which broadcasters and educators could improve
th e ir uses of rad io ta lk s as a means of imparting informa
tio n to the public. Broadcasters and educators should be
able to strengthen t h e i r c r i t e r i a of se lectio n with ex p eri
m entally derived knowledge of the r e la tiv e ex ten ts to which
the to p ic of an informative rad io ta lk is more or le ss im
p o rtan t than the p rese n ter of the ta lk in promoting in te re s t
and understanding among the audience.
The general problem of t h i s study was to measure the
e f f e c t s of in te r e s t in the topic and the ethos of the speak
er upon understanding of an informative rad io ta lk .
To t h i s end, the following n u ll hypotheses were
t e s t e d :
1. As measured by a r e te n tio n t e s t , understanding
of the ra d io ta lk on sleep fo r a c o n tro l group
of subjects who heard a physician did not d i f f e r
s ig n ific a n tly from th a t of those subjects who
lis te n e d to a newscaster.
2. The le v el of in te r e s t reported by those subjects
who liste n e d to the ta lk delivered by the sub
je c t-m a tte r s p e c ia lis t did not d i f f e r s i g n i f i
ca n tly from the le v e l of in te r e s t in the in fo r
mation reported by the subjects who heard the
newsman.
Among matched p airs of subjects who marked the
ta lk as "in te re stin g " ra th e r than "highly in te r
e s tin g ," understanding of the rad io ta lk on
sleep for the c o n tro l group of subjects who
heard the s p e c ia lis t did not d i f f e r s ig n i f i
cantly from th a t of the subjects who heard the
newscaster.
The average scores on the re te n tio n t e s t fo r the
subjects who marked the ta lk as "highly in t e r
e stin g " did not d i f f e r s ig n ific a n tly from the
scores made by th e ir matched co u n terp arts who
marked the ta lk as " in te r e s tin g ."
The le v el of in te r e s t in the inform ation re
ported by the group of male subjects did not d if
fe r s ig n ific a n tly from the average lev el of in
t e r e s t reported by the group of female su b je cts.
The average le v e l of in t e r e s t in the ra d io ta lk
indicated by the freshman subjects did not d i f
fer s ig n if ic a n tly from th a t of the sophomore
su b je cts.
Among subjects who were matched by ACE place
ment scores w ithin the same te n th of the d is
tr ib u tio n of scores on the ACE t e s t of scholas
t i c ap titu d e in Oregon S tate C ollege, the aver
age scores on the t e s t of understanding by those
89
whose ACE scores were below the six th te n th did
not d i f f e r s ig n if ic a n tly from those whose ACE
scores were in or above the s ix th te n th .
The examination of the l i t e r a t u r e revealed no study
d i r e c t l y r e la te d to the problem of t h i s in v e s tig a to r. Some
in d ire c t r e la tio n s h ip s were discovered in the s h if t- o f- o p in
ion stu d ie s of source c r e d i b i l i t y . The re p o r ts of Trenamen
were most d i r e c t l y r e la te d to the present study, but they
did not explore r e la tio n s h ip s of in te r e s t and a u th o rity to
re te n tio n of inform ation.
The few experim ental stu d ies th a t were intended to
measure l i s t e n e r responses to ra d io t a lk s were recorded
speeches th a t were played to large groups of su b je c ts . The
ta lk s were not p r e -te s te d fo r in d ic a tio n s th a t the l i s t e n e r
would a c tu a lly se le c t such a talk: on the b a s is of i n t e r e s t
i f he were not in a "captive" audience. This in v e s tig a to r
attem pted to create an experim ental procedure th a t would
avoid the lim ita tio n s in d ic ated here. The ra d io t a l k was
p re -te s te d fo r general i n t e r e s t ; the su b je c ts lis te n e d in
small inform al groups; the su b jects were placed in a s e ttin g
outside the classroom th a t approximated a normal lis te n i n g
s itu a tio n .
This experim enter d esire d a to p ic fo r a ra d io ta l k
th a t would meet five e s s e n t i a l requirem ents fo r purposes of
t h is study:
1. Both men and women should have general personal
90
i n te r e s t in the inform ation to be presented.
2. The topic should be somewhat fa m ilia r to the
su b je cts, and should be one on which they had
l i t t l e sp e cific knowledge but some opinion.
3. The topic should be appropriate as program mater
i a l for broadcasting as a short featu re ta lk in
a longer ra d io program.
4. The topic should lend i t s e l f to a proper t e s t
of re te n tio n of inform ation.
5. The topic should be acceptable as a statement
of an expert or as th a t,o f a person who is a
s k ille d read er.
Health and medical aspects of the topic of sleep
were selected for the study. This kind of inform ation ap
peared to meet the requirem ents of the in v e s tig a to r. This
experimenter spent several months in the com pilation of de
ta i l e d inform ation on the topic of sleep. From these data
fo rty items of inform ation were selected for incorporation
in to the rad io ta lk . This inform ation was examined by five
s ta f f members of the Speech Department of Oregon State Col
lege, and t h i r t y - f i v e items f in a ll y were selected fo r use in
preparing the radio s c r ip t . Two p ro fe ssio n al rad io w riters
of rad io s ta tio n KOAC, the sta te educational rad io s ta tio n
of Oregon, were consulted on the p reparation of the m aterial
and on the maximum length of the rad io ta lk to be incorpor
ated in to the program format.
91
Two radio s c rip ts were prepared; each was approxi
mately 1200 words in length and required a delivery-tim e of
seven to eig h t minutes. This length was needed for the pur
pose of introducing the ta lk as a p a rt of a long program.
A s ta f f member of the College who had p ro fessio n al
announcing experience recorded the two ta lk s under co n tro lled
engineering conditions. In order to p r e - te s t in te r e s t in
the t a l k , fo rty -e ig h t su b jects lis te n e d in small groups to
the two recorded ta lk s in a radio studio and marked th e ir
responses on a th ree -p o in t ra tin g scale of in te r e s t. Speech
#2 received a highly favorable r a tin g and was revised and
re-recorded by the same voice th a t made the previous record
ing. This ta lk then was te s te d for the in te r e s t of t h i r t y
a d d itio n a l su b je c ts, f if t e e n males and f if t e e n females. All
subjects indicated th a t the ta lk was e ith e r highly in t e r e s t
ing or in te r e s tin g . No subject found the ta lk d u ll. This
recording was then used fo r a l l fu rth e r experimental pro
cedures.
The experimental design required a vocal deliv ery
th a t would be equally acceptable to the lis te n e r as th a t of
a reputable physician or as th a t of a news commentator on a
large m etropolitan radio s ta tio n . Three s ta f f members of
Oregon S tate College who had p ro fessio n al rad io announcing
experience and d is tin c tiv e voices were selected for evalua
tio n by subjects like those who were to be in the f in a l ex
periment. Ninety ad d itio n a l su b jects, fo rty -fiv e men and
92
fo rty -fiv e women, were asked to id e n tify the speaker as
the physician or as the newscaster. A ll subjects were given
the th re e -p o in t ra tin g scale of in te r e s t in the ta lk . Thir
ty students ( f if te e n males and f if t e e n females) evaluated
each of the three recorded t a lk s . Voice #1 was highly sa t
isfa c to ry and was te s te d with t h i r t y a d d itio n a l su b jects.
The r e s u lt s strengthened the assumption th a t the voice was
acceptable to college subjects when i t was introduced as
the voice of e ith e r the physician or the newscaster.
T h irty -fiv e m ultiple-choice items were prepared for
the t e s t of understanding. A fter co n su lta tio n with s ta ff
members of the Speech and Psychology Departments in Oregon
State C ollege, tw enty-five of the items were te s te d with
eighty subjects in order to determine the discrim inative
■ i
strengths of the item. R esults of t h i s t e s t indicated th a t
a l l tw enty-five items should be re ta in e d .
An a d d itio n a l 120 subjects then heard the rad io ta lk
without p rio r id e n tif ic a tio n of the speaker. These subjects
reported th e ir grade-point averages of the previous term of
college study. The highest and lowest scores were compared
with grade-point average to determine if the t e s t were valid
in discrim in atin g between subjects who were matched fo r GPA.
A t - t e s t showed th a t students who reported high GPA ra tin g s
of the previous term 's study had higher scores than those
subjects who had low GPA's. This difference was s ig n ific a n t
at the 1 per cent le v el of confidence. The th re e -p o in t
93
scale of in te r e s t also was submitted to these su b jects.
Seventy-four subjects marked the ta lk as in te re s tin g and
f o rty -s ix subjects found the ta lk highly in te re s tin g . No
student reported th a t the ta lk was d u ll.
The radio ta lk was then incorporated in a longer
program format and introduced as p a rt of a "Family Hour"
from a large m etropolitan rad io s ta tio n in Denver, Colorado.
Each in tro d u ctio n of the speaker was prepared in co n su lta
tio n with p ro fe ssio n a l w riters of rad io s ta tio n KOAC; f i n a l
passages of standard song preceded each in tro d u ctio n of the
speaker and followed the end of the t a l k .
The eighty subjects of the f i n a l p art of the study
were matched for age, sex, year in c o lle g e , previous term 's
GAP, and ACE placement score. S ubjects in groups of five
heard the rad io ta lk in the rad io studio under co n tro lle d
co n d itio n s. Forty subjects heard the speaker introduced as
the physician and fo rty subjects heard the speaker in tr o
duced as the newscaster. All su b jects were given the t e s t
for re te n tio n of inform ation and the th re e -p o in t ra tin g
scale of in t e r e s t .
Thus, two s e ts of measurements were av ailab le for
determining the e f f e c ts of the rad io ta lk upon the subjects:
(1) the ra tin g s of the in te r e s t value of the t a l k , and (2)
the scores made by the subjects on the re te n tio n t e s t . In
processing these d ata, three p rin c ip a l s t a t i s t i c a l proced
ures were used: (1) the t - t e s t , (2) c h i square, and (3)
94
an a ly sis of variance.
Primary Conclusions
As previously rep o rted , the general problem of th i s
study was subdivided into seven c o n s titu e n t p a r ts , each ex
pressed as a hypothesis, which were te s te d . In lig h t of
these t e s t s and w ithin the lim ita tio n s of t h i s experim ental
design, the following conclusions seemed j u s t i f i e d :
1. As measured by a re te n tio n t e s t , understanding
of a radio ta lk on sleep by subjects who thought
th a t they were lis te n in g to a physician was sig
n ific a n tly g re ater than the understanding of
subjects who thought they were lis te n in g to a
newscaster.
2. The le v e l of in te r e s t reported by subjects who
thought th a t they were lis te n in g to a physician
was s ig n ific a n tly higher than the in te r e s t le v e l
reported by subjects who thought they were l i s
tening to a newscaster.
3. Among matched p a irs of su b jects who marked the
ta lk as "in te re stin g " ra th e r than "highly in te r
e s tin g ," subjects who thought th a t they were
lis te n in g to a physician made s ig n ific a n tly
higher scores on the re te n tio n t e s t than sub
j e c t s who thought th a t they were lis te n in g to a
rad io newscaster.
95
4. The average scores on the re te n tio n t e s t for the
subjects who marked the ta lk as "highly in te r
estin g " did not d if f e r s ig n ific a n tly from the
scores made by t h e i r matched co u n terp arts who
marked the ta lk as " in te r e s tin g ."
5. The lev el of in te r e s t in the inform ation r e
ported by the group of male subjects did not dif
f e r s ig n ific a n tly from the average le v e l of in
t e r e s t reported by the group of female su b jects.
6. The average le v el of in te r e s t in the ra d io ta lk
indicated by the freshman su b jects did not d if
fe r s ig n ific a n tly from th a t of the sophomore
su b jects.
7. Subjects with high ACE r a tin g s had s ig n ific a n tly
higher scores on the inform ation t e s t than did
those with low ACE r a tin g s .
Secondary Conclusions
Within the lim ita tio n s of t h i s study, c e rta in con
clu sio n s may be drawn th a t are of significance to the educa
to r or to the rad io broadcaster planning inform ational radio
t a lk s . Is the experience of the speaker on the topic impor
ta n t to the l i s t e n e r ’s understanding of the informative
ta lk s th a t are presented through rad io broadcasting statio n s?
Is the medium of rad io so w e ll-e stab lish ed in the public
mind th a t the channel of communication i t s e l f is s u f fic ie n t
96
to work acceptance upon the lis te n e r ? The r e s u l t s of t h i s
experiment in d icated th a t the apparent ex p ertn ess of the
speaker is of s ig n ific a n t importance in determ ining the
amount of understanding of the inform ation presented to the
l i s t e n e r . When the two matched groups of su b je cts who l i s
tened to the physician and the news commentator were com
pared, those su b jects who heard the e x p e rt, Dr. Maloney, had
s ig n if ic a n tly g re a te r understanding of the ra d io ta l k than
the su b je cts who lis te n e d to the news commentator, Chet Par
ker. On the b a sis of r e te n tio n of inform ation, the s ta tu s
of the expert appeared to r e s u l t in more a tte n tiv e l i s t e n
ing to the inform ation than the s ta tu s of the news commenta
to r who rep resen ted the broadcasting in d u stry . The accept
ance and understanding of the h e a lth and medical asp ects of
sleep were g re a te r fo r the expert on the to p ic . This con
c lu sio n supported the assumption of those educators and
b ro ad casters who p re fe r to use an expert fo r the d e liv e ry of
an inform ative rad io ta l k in stead of a s ta f f announcer. Ad
d itio n a l rese arch would be needed using a v a rie ty of to p ic s
to determine those to p ic s on which the broadcaster would be
as acceptable as the in f o rm a tio n -s p e c ia lis t.
Is in te r e s t in the topic of the ta lk more important
than the a u th o rity of the speaker? Is the ex ten t of i n t e r
e s t r e la te d d ir e c tly to the a u th o rity or expertness of the
speaker on the topic of an inform ative rad io ta lk ? This
study attem pted to measure the r e la tiv e importance of in te r -
97
e s t in the topic when the two types of a u th o rity delivered
the same inform ation. The group of su b jects who heard the
physician re g iste re d a s ig n ific a n tly higher le v el of in te r
e s t in the rad io ta lk than those su b jects who heard the news
commentator. Was t h i s the r e s u lt of the influence of the
expert or the r e s u lt of a strong p re d isp o sitio n in favor of
the inform ation by one group of subjects? Did le v el of in
t e r e s t ra th e r than influence of the speaker determine the
d iffe re n ce s in understanding of the inform ation? The
matched p a irs of subjects who marked the ta lk as in te re s tin g
ra th e r than highly in te re s tin g were compared. The subjects
who heard the s p e c ia lis t had a s ig n ific a n tly higher average
understanding of the rad io inform ation than the group who
heard the newscaster. The conclusion was th a t the d i f f e r
ence in understanding between the two groups was due to the
influence of the speaker, since both groups indicated the
same average lev el of in te r e s t in the to p ic .
To in v e stig a te fu rth e r the influence of in te r e s t in
re la tio n s h ip to the speaker source, the subjects who re
ported th a t the ta lk was highly in te re s tin g were compared
with th e ir matched co u n terp arts who in d icated th a t the ta lk
was in te re s tin g . Both groups had the same average scores on
the t e s t of inform ation. I t may be concluded th a t in te re s t
in the topic was not s u ffic ie n t to account for the d i f f e r
ences in understanding th a t were re g is te re d by the two
groups of subjects who heard the physician and the news
98
c a s te r . Within the lim its of t h i s study, i t may be con
cluded th a t the acceptance of the source of rad io informa
tio n is more important than the degree of i n t e r e s t evidenced
in the topic in determ ining the amount of inform ation under
stood by the l i s t e n e r . This conclusion tends to r e f u te the
point of view of some b ro ad casters and educators th a t the
amount of in t e r e s t in an inform ative ra d io ta lk shown by
the l i s t e n e r is the most important element in gaining maxi
mum communicative e f f e c t s from the p re s e n ta tio n .
The male su b je c ts in the experiment in d ic ated the
same average le v e l of i n t e r e s t in the ta lk as the female
su b je c ts . I t may be concluded th a t the f a c to r of sex did
not operate as a s ig n if ic a n t v ariab le in the r e s u l t s of t h i s
study. The inform ation on the topic of the ra d io t a l k was
of general in te r e s t to both male and female su b je c ts.
The su b je cts who were used in t h i s experiment were
equally divided between freshmen and sophomores. This ex
perim enter f e l t th a t by using the two groups a broader sam
ple would be a v a ila b le fo r comparison in term s of age and
m a tu rity . However, the freshmen su b je c ts used in t h i s ex
periment in d ic ated the same average le v e l of in t e r e s t in the
rad io t a l k as the sophomore su b je cts. Age and year in co l
lege did not operate as s ig n ific a n t v a ria b le s in the r e s u l t s
of t h i s in v e s tig a tio n . I t may be concluded th a t the se
lected inform ation used on the topic of sleep was of appro
p ria te in t e r e s t to both freshmen and sophomores.
99
Those su b jects in the experimental and co n tro l
groups who had ACE r a tin g s in the sixth te n th or above
earned s ig n ific a n tly higher scores on the r e te n tio n t e s t
than those su b jects who had ACE ra tin g s below the sixth
te n th . This comparison was a t e s t of the discrim inatory
power of the re te n tio n items as measures of understanding.
If the more in te lli g e n t students as measured by ACE scores
had not made s ig n ific a n tly higher scores in the r e te n tio n
t e s t , the t e s t or the method of matching would be highly
suspect. The r e s u l t s were c o n siste n t with r e s u l t s of the
comparisons th a t were made during prelim inary procedures
among groups of students who were assumed to be ty p ic a l of
the subjects who were in the f i n a l experiment.
Im plications
In the experim ental conditions under which th i s
study was made, the variable of the id e n tity of the speaker
appeared to a ffe c t s ig n ific a n tly the understanding of the
radio ta l k , despite the strong personal in te r e s t in the in
formation th a t was reported by both the experim ental group
and the c o n tro l group. Although in te r e s t in a topic is a
prelim inary m otivation for the lis te n e r to tune to a radio
program, an in te re s tin g rad io ta lk th a t is designed to in
form may be more e ffe c tiv e in i t s purpose i f , in ad d itio n to
presenting a topic of general in t e r e s t , the speaker is r e
garded as being highly a u th o rita tiv e on th a t to p ic .
100
This finding tends to confirm the conclusion of
Trenaman in h is studies of B ritis h ta lk s th a t a high le v e l
of in te r e s t in a topic does not n ec essarily r e s u l t in g re a t
er understanding of the inform ation p re se n te d .- ^ In h is
study of inform ation presented by te le v is io n , Brandon also
found no s ig n ific a n t c o rre la tio n between le v e l of in te r e s t
th a t was expressed by the viewer and achievement on a t e s t
2
of inform ation in the program.
Hoviand, Lumsdaine, and S h effie ld in t h e i r in v e s ti
gations of imparting inform ation by film found no d ir e c t re
la tio n sh ip between understanding and i n t e r e s t . They con
cluded:
I t i s c le a r th a t the amount of in te r e s t or extent
of approval provides only supplementary inform ation
about a film with an education purpose and does not
c o n s titu te a measure of i t s e ffe c tiv e n e ss at achieving
i t s educational o b je c tiv e s .3
On the b asis of the present study, i t is possible to
in fer th a t broadcasters have placed too much emphasis on
in te re s t in the topic as an end in i t s e l f in assessing the
success of a radio program th a t is designed to impart in-
\loseph Trenaman, "Understanding Radio T alks," The
Q uarterly Journal of Speech. XXXVII (A pril, 1951), 178.
2
James R. Brandon, "An Experimental T elevision
Study: The R elative E ffectiveness of Presenting Factual
M aterial by the Lecture, Interview , and D iscussion Methods,"
Speech Monographs. XXIII (November, 1956), 282.
^Carl I. Hovland, Arthur Lumsdaine, and Fred Shef
f i e l d , Experiments on Mass Communication. Vol. I l l : Studies
in S ocial Psychology in World War I I (4 v o ls.; Princeton:
Princeton U niversity Press, 1949-50), pp. 230-35.
101
formation to the l i s t e n e r . The id e n tity of the speaker may
be even more important in stim ulating th a t le v e l of in ten
sive lis te n in g e s s e n tia l for understanding and accepting
the inform ation.
Another im plication suggested by the r e s u l t s of t h i s
in v e stig a tio n i s the p o s s ib ility th a t for conveying some
kinds of information with maximal informative r e s u l t s , the
radio newscaster or p ro fe ssio n a l radio announcer is the much
poorer instrument than the inf or mat i o n - s p e c i a l i s t . I t may
be suggested th a t the more meaningful the topic of the radio
ta lk to the immediate and personal welfare of the lis te n e r
and h is fam ily, the g re a te r are the e f f e c ts of the presenta
tio n of the inform ation by the in f or mat ion-spec ia l i s t , as
co n trasted with the p ro fessio n al newscaster or commentator,
in heightening in te r e s t and understanding among the members
of the audience.
Recommendations
Further research is necessary to determine the top
ics th a t can be presented most e ff e c tiv e ly by the informa
tio n - spec ia l i s t as the speaker, and those su b ject-areas
which can be reported with equal or g reater e ffe c tiv e n e ss by
the p ro fe ssio n a l radio jo u r n a li s t.
The se le c tio n of the two sources of tru stw o rth in ess
th a t were in v e stig a ted may have had an e ffe c t on the li s t e n
er th a t would not be c o n siste n t with the r e s u l t s of comparing
102
other types of a u th o rity . Perhaps a physician has a sp e cial
p restig e th a t is not comparable to th a t of the inform ation-
s p e c ia lis t in other f ie ld s such as a g r ic u ltu r e , home econom
ic s , or psychology. Further research with other types of
a u th o rity should be pursued before any general conclusions
may be drawn safely .
In preparing the rad io ta lk fo r t h i s experiment,
care was taken to avoid any documentation of the informa
tio n in the s c rip t. There were no quotations or referen ces
to sources. If the inform ation were documented, would the
newscaster and the physician be equally e ffe c tiv e in arous
ing in te r e s t and imparting inform ation to the lis te n e r ?
This is a problem of future research . C athcart studied the
use of documented evidence as re la te d to s h if t of opinion on
the topic of the a b o litio n of c a p ita l punishment. He con
cluded th a t documenting the sources of inform ation had no
s ig n ific a n t e ffe c t on the favorable acceptance of argument
as measured by s h if t of opinion.4 No rep o rt has appeared in
the l it e r a tu r e on the e ffe c t of documentation of inform ation
in a radio ta lk upon source c r e d i b i l i t y .
R esults of the comparison may have been d iffe re n t
had the newscaster in t h i s study been a network p erso n a lity .
If a popular radio or te le v is io n c e le b rity or an outstanding
4Robert S. C athcart, "An Experimental Study of the
R elative E ffectivenss of Four Methods of Presenting Evi
dence," Speech Monographs, XXII (August, 1955), 233.
103
j o u r n a l i s t who had p re stig e of n atio n a l renown were to tu rn
to sp e cia lize d to p ic s , would h is e ffe c tiv e n e s s compare fa
vorably with th a t of a s p e c ia lis t on the to p ic ? There is
s u b s ta n tia l evidence to in d icate th a t the popular c e le b r ity
can move l i s t e n e r s to a c tio n in p a t r io t ic and c h a rita b le
ap p eals, but none to in d ic a te how acceptable an Arthur God
fre y or an Edward R. Murrow would be in dissem inating speci
a liz e d inform ation a ffe c tin g the h e a lth or personal welfare
of the l i s t e n e r . At what point does the l is te n e r no longer
regard the c e le b r ity or the rad io j o u r n a li s t as a t r u s t
worthy source of inform ation? These questions are d i f f i c u l t
to study in a la b o ra to ry s e ttin g . New experim ental designs
are needed to evaluate in depth the importance of source-
c r e d i b i l i t y in network broadcasting.
The r e la tio n s h ip of in te r e s t in d if f e r e n t to p ic s to
various c h a r a c t e r i s t i c s of ra d io lis te n in g has been explored
e x ten siv e ly in the p re p a ra tio n of commercial announcements,
but i t has received l i t t l e study by e i th e r commercial broad
c a s te r s or ed u catio n al b ro ad ca ste rs in c re a tio n of the in
formative rad io t a l k . The rad io t a l k often is regarded by
the commercial b ro ad caster as a p art of the rad io program
schedule th a t has l i t t l e commercial value because i t s audi
ences are u su a lly sm all. Educational b ro ad ca ste rs have had
Robert K. Merton, Mass Persuasion (New York: Harper
& B rothers, 1946), p. 3.
104
n eith er the personnel nor the fin a n c ia l resources for deep
studies of audience responses to ta lk s . I t is the sincere
hope of t h i s in v e stig a to r th a t the experiment reported in
t h i s paper may stim ulate fu rth e r research on informative
radio t a l k s . As the la te President of the National Associa
tio n of Broadcasters sta te d :
In the f i n a l a n a ly s is , the most c r i t i c a l l y important
challenge th a t broadcasting has is to inform people; I
do not mean to educate the people in a formal sense, but
to inform them through the reg u la r and accurate re p o rt
ing of cu rren t events, through the scheduling of discus
sions th a t w ill provoke thought and thus illum inate de
c is io n , and through the p rese n tatio n of sp ecial fe a tu re s
in a l l of the a r ts and sciences th a t w ill make the great
lessons of lif e palatable and understandable to a l l of
u s .°
^Harold E. Fellows, “Memorandum to a Student of
B roadcastinq," Journal of B roadcastinq. I l l (Summer, 1959).
212 . . !
B I B L I O G R A P H Y
BIBLIOGRAPHY
Books
B ritish B roadcasting* London: B ritis h Broadcasting Corpora
tio n , 1954.
C hester, Giraud, and G arrison, Garnet R. T elevision and
Radio. 2nd ed. New York: Appleton-Century-Crofts,
In c ., 1956.
Cooper, Lane. The Rhetoric of A r i s t o t l e . New York: Apple-
ton-Century-CroftsV Inc., 1932.
Cronbach, Lee J . E sse n tials of Psychological T e stin g . New
York: Harper and B rothers, 1949.
G a rre tt, Henry E. S t a t i s t i c s in Psychology and Education.
5th ed. New York: McGraw-Hill Book Company, In c .,
1958.
G uilford, J . P. Fundamental S t a t i s t i c s in Psychology and
Education! 2nd ed. New York: McGraw-Hill Book Com-
pany, I n c ., 1950.
Head, Sydney W . Broadcastinq in America. Boston: Houghton
M ifflin Company’, 1956.
Hovland, C. I . , Lumsdaine, A. A., and S h effie ld , Fred D.
Experiments on Mass Communication. Vol. I l l : St u ri
le s in S ocial Psychology in World War I I . Prince
ton: Princeton U niversity Press, 1949-50.
Lindzey, Gardner (e d .). Handbook of Social Psychology. 2
vols. Cambridge: Addison-Wesley Publishing" Company,
1954.
Merton, Robert K. Mass Persuasion. New York: Harper and
B rothers, 1946.
Schramm, Wilbur (e d .). The Process and E ffects of Mass
Communication. Urbana; U niversity of I l l i n o i s
Press, 1954.
106
107
Whitney, Frederick Lamson, The Elements of Research. 3rd
ed. New York: Prent'i'c e-H all', IncY, 1950.
W illis , Edgar E. Foundations in B roadcasting. New York:
Oxford U n iv ersity P ress, I n c ., 1951.
P e rio d ic a ls and T ests
American Council on Education Psychological Examination fo r
College Freshmen" 1954 ed". fYinceton: Educational
T esting Service, 1954.
Brandon, James R. "An Experimental T ele v isio n Study: The
R elative E ffectiveness of Presenting F actu al Mater
i a l by the L ecture, Interview , and D iscussion
Methods," Speech Monographs, XXIII (November, 1956),
272-83.
C a r t ie r , F rancis A., and Harwood, Kenneth A. "Some Ques
tio n s about A tte n tio n ," The Jo u rn al of Communica
t i o n . . VIII (Autumn, 195&) , ~ l0 6 - 1(Y "
C a th c a rt, Robert S. "An Experimental Study of the R elative
E ffectiv en ess of Four Methods of Presenting Evi
dence," Speech Monographs. XXII (August, 1955).
227-33.
D ie tric h , John E. "The R elative E ffectiv e n ess of Two Modes
of Radio D elivery in Influencing A ttitu d e s ," Speech
Monographs. X III (No. 1, 1946), 58-65.
Ewbank, Henry Lee. "Trends in Research in Radio Speech,"
The Q uarterly Jo u rn al of Speech, XXVI (A pril, 1940),
282-87.
Fellows, Harold E. "Memorandum to a Student of Broadcast
ing," Journal of B roadcasting. I l l (Summer, 1959),
208-14.
Haiman, Franklyn S. "An Experimental Study of the E ffec ts
of Ethos in Public Speaking," Speech Monographs.
XVI (September, 1949), 190-202.
Highlander, John P. "Audience Analyzer Measurements and
Inform ational E ffec ts of Speaker V ariables in Radio
T alks," Speech Monographs. XXI (August, 1954), 188-
89.
108
Kelman, Herbert C ., and Hovland, C arl I. 1 1 'R einstatem ent1
of the Communicator in Delayed Measurement of Opin
ion Change," Journal of Abnormal and S ocial Psy
chology. XLVIII I J u ly , 1953J , 327-35.
Knower, F ranklin H. "Experimental Studies in Changes in
A ttitu d e s: I. A Study of the Effect of Oral Argument
on Changes, of A ttitu d e ," Journal of S ocial Psychol
ogy, VI (August, 1935), 315-45.
Lindgren, Henry C. "The Effect of the Group on the Behav
ior of the In d iv id u al," E d u catio n ,' LXXIII (February,
1953), 382-88.
NAEB J o u rn a l. X-XVII (1950-58).
P arrish , Jack A., and Campbell, Donald T. "Measuring Prop-
agandat E ffects with D irect and In d irect A ttitude
T e sts, Journal of Abnormal and S ocial Psychology,
XLVIII (January, 1953), 3-9.
Paulson, Stanley F. "The E ffects of the P restige of the
Speaker and Acknowledgement of Opposing Arguments
on Audience R etention and S h ift of Opinion," Speech
Monographs, XXI (November, 1954), 267-71.
Psychological A b s tra c ts . XIV-XXXII (1940-58).
Schank, R. L ., and Goodman, C. "Reactions to Propaganda on
Both Sides of a C o n tro v ersial Issu e," Public Opinion
Q u a rte rly . I l l (1939), 107-12.
Sponberg, Harold. "The R elative E ffectiveness of Climax
and Anti-Climax Order in an Argumentative Speech,"
Speech Monographs, XIII (no. 1, 1946), 35-44.
"Summary of Commercial Broadcasting," B roadcasting, LVII
(September 14, 1959), 133.
The Journal of Communication. I-V III (1951-1958).
"Theses and D is s e rta tio n s in Broadcasting." Journal of
Broadcasting, II (Winter, 1957-58), 57-90.
Townsend, Howard W . "Factors of Influence in Radio Speech,"
The Q uarterly Journal of Speech. X X X (A pril, 1944),
187-90.
Trenaman, Joseph. "Understanding Radio T alks," The Quarte r
ly Journal of Speech, XXXVII (A pril, 1951), 173-78.
109
Whan, F orest L. "The Speech Profession J i l t s Radio," The
Q uarterly Jo u rn al of Speech, XXX (December, 1944),
439-44.
W illis , Edgar E. "Research in Radio and T elevision by Grad
uate Students in Speech," The Q uarterly Jo u rn al of
Speech, XLI (October, 1955), 261-70.
A P P E N D I X E S
APPENDIX A
PRELIMINARY TEST FOR DETERMINING SCORES
BASED ON PRIOR KNOW LEDGE AND CHANCE
Please follow these in s tru c tio n s :
This t e s t is part of an experiment. Please tr y to
answer a l l of the questions on the b asis of your present
knowledge and your personal opinion. TRY TO M AKE SOME KIND
OF RESPONSE TO EVERY QUESTION. You need not place your name
on your paper. Your cooperation on t h i s experiment is
deeply appreciated.
CIRCLE YOUR CHOICE FOR THE CORRECT ANSW ER TO
EACH QUESTION.
1. Last year we Americans purchased a quantity of sleep
ing p i l l s and powders equal to (a) six tons (b) two
tons (c) four tons.
2. The id eal bedroom temperature is about (a) 55 degrees
(b) 68 degrees ( c ) 75 degrees.
3. In the United S ta te s, the occupation group which aver
ages the la rg e s t amount of time spent in sleeping is
(a) Schoolteachers (b) Policemen (c) Bank T e lle rs .
4. For most e f f i c i e n t sleep, the individual sleeper should
use a bed (a) 36 inches wide (b) 50 inches wide (c) 75
inches wide.
5. The most r e s t f u l bedroom c o lo rs are (a) blues and
greens (b) pinks and yellows (c) browns and grays.
6. The a b i l i t y to get to sleep is most a ffected by (a) a
cold shower p rio r to r e t i r i n g (b) two cups of coffee
p rio r to r e t i r i n g (c) a bowl of hot soup p rio r to re
t i r i n g .
7. The f i r s t u n iv e rsity to introduce a course in Sleep
was (a) U niversity of Iowa (b) U niversity of Texas (c)
U niversity of C a lifo rn ia .
8. The average human being who goes without any sleep con
tin u a lly w ill die w ithin (a) four days (b) seven days
(c) twelve days.
9. Fatigue is caused by (a) an accumulation of la c tic acid
in the muscle tis s u e (b) an increase in blood pressure
(c) an increase in metabolism.
112
113
10. Americans sleep more (a) in the month of December (b)
in the month of A pril (c) in the month of July than
in any other month.
11. For adequate r e s t , most in d iv id u als require (a) four
hours of deep sleep (b) six hours of deep sleep (c)
three hours of deep sleep.
12. In ch ild re n , the depth of sleep can be measured (a) by
decrease in blood pressure (b) by increase in the
amount of p e rsp ira tio n (c) by increase in blood pres
sure .
13. The worst method of sleeping e f f i c i e n t l y is (a) sleep
ing curled up in a b a ll (b) sleeping with one arm.
under your pillow (c) sleeping on your back.
14. C hildren should d ic ta te th e ir own needed sleep (a) up
to the age of four (b) up to the age of six (c) up to
the age of two.
15. In the Spring months, we dream (a) twenty per cent
more (b) fo rty per cent more (c) f i f t y per cent more
than during the Winter months.
16. A fter changing your reg u lar sleep ro u tin e , i t u su a lly
re q u ire s (a) three nights (b) fiv e nights (c) seven
nights to re a d ju st completely to your new ro u tin e .
17. If you cut your needed amount of sleep ton ig h t by 50/6,
to complete the same amount of work tomorrow th a t you
o rd in a rily do, you w ill require (a) 40/6 more energy
(b) 60/a more energy (c) 100/o more energy.
18. In terms of natio n al h ab its among c o u n trie s, the most
e f f i c i e n t sleepers are (a) the Mexicans (b) the B ritis h
(c) the Germans.
19. Less than (a) 20/6 (b) 4 0 /b (c) 60/6 of married couples
sleep in single beds.
20. Business executives need (h alf as much sleep (b) twice
as much sleep (c) about one-third as much sleep as the
average manual lab o re r.
21. A short nap in the daytime has (a) no e f f e c t on one's
a b i l i t y to get to sleep a t night (b) great e f f e c t on
one's a b i l i t y to get to sleep at night.
22. Nervous, ex citab le persons (a) need more sleep and get
114
le s s (b) need le s s sleep and get more (c) need more
sleep and get more than more p la c id , calm in d iv id u a ls .
23. The soundest sleep for the average ad u lt is (a) between
the t h ir d and fo u rth hours (b) between the f i f t h and
s ix th hours ( c ) between the second and th i r d hours.
24. (a) About 50)6 (b) about 60/6 (c) about 7 0 & of us Ameri
cans wear pajamas to bed.
25. Hypnosis resembles sleep in th a t (a) the co rte x of the
b ra in is in h ib ite d (b) the muscle system is completely
relax ed (c) the blood pressure i s increased.
PLEASE M AKE CERTAIN THAT ALL QUESTIONS HAVE BEEN
ATTEMPTED: THANK YOU FOR YOUR ASSISTANCE IN THE
EXPERIMENT .
APPENDIX B
RADIO TALK USED IN THIS STUDY
RADIO TALK
W e Americans are notoriously the w orld's most inad
equate sleep ers. The B r itis h are regarded as the most e f
f ic ie n t in terms of reg u lar hours and proper slumber h a b its .
You know, lack of sleep has become one of America's most
serious health hazards. Last year we purchased more than
four tons of sleeping p i l l s and powders in order to b a ttle
insomnia. A ll of us have been to ld at some time th a t the
three e s s e n ti a ls for lif e are food, s h e lte r , and clo th in g .
The fo u rth e s s e n tia l is sleep, and is equally important. As
a matter of f a c t , you can live fa r longer without food than
without sleep. A fter four days without any sleep a t a l l ,
the brain c o n tro ls begin to d is in te g ra te and a f te r an ex
treme lim it of seven days without any sleep, the individual
w ill in e v ita b ly d ie. The human being, the dog, and the mon
key have about the same l i f e lim it in terms of liv in g with
out sleep, seven days.
Despite the importance of sleep, most of us have
never studied the subject very much and t h i s p a r tia lly ac
counts for our national insomnia. The U niversity of Texas
was the f i r s t college to introduce a course on sleep to im
prove freshmen grades, and eleven other schools are now of-
116
117
ferin g sim ilar courses. L e t's examine some of the informa
tio n on sleep th a t might a f f e c t your h ealth and mine.
Fatigue is caused by an accumulation of la c tic acid
in the muscle tis s u e . When t h i s la c tic acid accumulates to
a c e rta in point, the body re q u ire s a period of sleep to d is
sipate t h i s oversupply of la c tic acid. Sleep, th e re fo re ,
may be regarded as a remedy or as the mechanic who re a d ju s ts
th ','hanism. Also during sleep, the cortex of the
br Ib ited and the muscles become relax ed . In hyp-
nc tex of the b rain is also in h ib ite d , but hyp-
n o u o cannot replace n atu ral sleep in re p a irin g the ravages
of fa tig u e .
Now most people believe th a t sleep re q u ire s eight
hours in bed. However, you can re s t ju s t as e a s ily in a
comfortable c h a ir in a relaxed s it t i n g p o sitio n , and awaken
j u s t as refre sh ed . Changing your sleeping ro u tin e , however,
re q u ire s about three nights fo r readjustm ent. If p o ssib le ,
you should not vary such h a b its as using a pillow , sleeping
in pajamas, or changing bedrooms without expecting some lo ss
in sleep e ffic ie n c y fo r about three n ig h ts. Most people
require four hours of so -called "deep" sleep, but t h i s may
a c tu a lly require eight or ten hours in bed. The depth of
sleep can be measured in c h ild ren by the amount of p ersp ira
tio n . Small ch ild re n should perspire fre e ly when sleeping
soundly. As y et, the depth of sleep has never been measured
adequately for the a d u lt. It is estim ated t h a t , fo r a d u lts ,
118'
the soundest sleep i s usually during the th ir d and fo u rth
hours.
You know, babies and e ld e rly people do not a c tu a lly
sleep much more in twenty-four hours than the average ad u lt,
despite opinion to the c o n tra ry . They do req u ire more short
periods of sleep with naps during the day but the t o t a l
sleep they normally require is not much more than is needed
by the average middle-aged a d u lt.
The nervous, ex c ita b le person needs more sleep and
gets le ss than the calmer in d iv id u al. Up to the age of six,
a c h ild should d ic ta te h is own sleep h ab its in terms of the
number of hours. Some ch ild re n sleep very soundly and need
le s s time in bed than other ch ild re n to get the equivalent
amount of r e s t . The weather has a s ig n ific a n t e f f e c t on our
slumber h a b its. W e dream f i f t y per cent more in the spring
than in the w inter. N inety-five per cent of us go to sleep
more e a s ily in the spring, and move le ss during sleep. In
w inter, we average about seventy-five major movements during
eig h t hours of r e s t ; in the spring, about sixty major move
ments.
During sleep, the blood pressure drops and the mus
c le s become relax ed . You may sleep in any p o sitio n , but
never c u rl up in a b a l l . This p o sitio n prevents any muscu
la r re la x a tio n . A short nap during the day has no e ffe c t
on one's a b i l i t y to go to sleep at night. Eating a large
meal a t dinnertime also has no e f f e c t on sleep, unless you
119
have in d ig e stio n problems. You may even drink two cups of
coffee or have a bowl of hot soup a t bedtime if you f e e l
hungry. Neither w ill keep you awake physiologically; howev
e r , you may think yourself into sleeplessness by believing
th a t the coffee or soup w ill keep you awake. Never sleep in
a cold bedroom if you desire complete r e s t . The ideal bed
room temperature i s 68 degrees with proper v e n tila tio n . If
you want complete r e s t , never set an alarm clock to awaken
you. I t has a s ig n ific a n t e f f e c t on the soundness of your
r e s t . W e sleep more in A pril than in any other month.
Mental fatigue req u ire s more sleep to re p a ir damage
than physical fa tig u e . The business executive a c tu a lly
req u ires twice as much sleep as the manual lab o re r, on the
average. Young people tend to ignore the demands of fatigue
as long as p o ssib le. However, a warning. If you cut your
needed sleep f i f t y per ce n t, on the next day i t w ill require
fo rty per cent more energy to accomplish the same amount of
work you could have done a f te r a f u l l n ig h t's r e s t . And
mistakes can be c o s tly , which is perhaps one reason why bank
t e l l e r s as a group spend more hours in sleeping than any
other occupational group.
Noises are deadly for sound sleep. A noisy room may
not awaken you, once you are in deep sleep, but, neverthe
le ss, the noise w ill ra is e your blood pressure and prevent
f u l l slumber b e n e fit. The ideal bedroom w ill be decorated
in blues or greens, the two most r e s t f u l co lo rs. A ll lig h t
and a l l noise should be elim inated. Never take a cold show
er before r e t i r i n g or exercise heavily. Take a warm shower
if p o ssib le , and do not take your problems to bed to solve
mentally during the slumber period. Id e a lly , everyone
should sleep alone for f u l l r e s t; a c tu a lly le ss than fo rty
per cent of married couples sleep in single beds, and the
size of the bed is very important. For muscular r e la x a tio n ,
the bed should be at le a s t f i f t y inches wide. C oil springs
save about f i f t e e n per cent more energy than link springs,
and the springs should be f a i r l y firm . The dormitory beds
used in some h o sp ita ls and college dorm itories are fa r too
narrow for best slumber r e s u l t s .
Men need more sleep but a c tu a lly get le ss than
women, for a v a rie ty of reasons. Never tuck in the covers
at the sides or the foot of the bed. Covers should be lig h t
in weight so as to leave as l i t t l e pressure on the body as
p o ssib le. About six ty per cent of us Americans wear pajamas
to bed. For complete re la x a tio n of a i l muscles, id e a lly we
should wear no night clo th in g at a l l . The three great heal
e rs are food, water, and sleep. If you miss a n ig h t's r e s t ,
i t w ill take a week or more to make up t h i s loss com pletely.
Have you checked up on your slumber h ab its recently?.
APPENDIX C
INTEREST RATING SCALE AND
RETENTION TEST
(M ARK ONE) The ta lk th a t you ju s t heard was:__ 1._INTER
ESTING_______, 2. HIGHLY INTERESTING , 3.
DULL
CIRCLE YOUR CHOICE FOR THE CORRECT ANSW ER TO EACH
QUESTION
1. Last year we Americans purchased a quantity of sleep
ing p i l l s and powders equal to (a) six tons (b) two
tons ( c ) four tons.
2. The ideal bedroom temperature is about (a) 55 degrees
(b) 68 degrees (c) 75 degrees.
3. In the United S ta te s , the occupation group which aver
ages the la rg e st amount of time spent in sleeping is
(a) Schoolteachers (b) Policemen (c) Bank T e lle r s .
4. For most e f f ic ie n t sleep, the individual sleeper should
use a bed (a) 36 inches wide (b) 50 inches wide (c)
75 inches wide.
5. The most r e s t f u l bedroom c o lo rs are (a) blues and
greens (b) pinks and yellows (c) browns and grays.
6. The a b i l i t y to get to sleep is most affe c te d by (a) a
cold shower p rio r to r e t i r i n g (b) two cups of coffee
p rio r to r e t i r i n g (c) a bowl of hot soup p rio r to re
t i r i n g .
7. The f i r s t u n iv e rsity to introduce a course in Sleep
was (a) U niversity of Iowa (b) U niversity of Texas (c)
U niversity of C a lifo rn ia .
8. The average human being who goes without any sleep con
tin u a lly w ill die w ithin (a) foui days (b) seven days
(c) 12 days.
9. Fatigue is caused by (a) an accumulation of la c tic acid
in the muscle tis s u e (b) an increase in blood pressure
(c) an increase in metabolism.
10. Americans sleep more (a) in the month of December (b)
in the month of A pril (c) in the month of July than in
any other month.
11. For adequate r e s t , most individuals require (a) four
hours of deep sleep (b) six hours of deep sleep (c)
122
123
three hours of deep sleep.
12. in c h ild re n , the depth of sleep can be measured (a)
by decrease in blood pressure (b) by increase in the
amount of p e rs p ira tio n (c) by increase in blood pres
sure .
13. The worst method of sleeping e f f i c i e n t l y is (a) sleep
ing curled up in a b a l l (b) sleeping with one arm under
your pillow (c) sleeping on your back.
14. C hildren should d ic ta te th e ir own needed amount of
sleep (a) up to the age of four (b) up to the age of
six (c) up to the age of two.
15. In the Spring months, we dream (a) twenty per cent more
(b) fo rty per cent more (c) f i f t y per cent more than
during the Winter months.
16. A fter changing your re g u la r sleep ro u tin e , i t u su ally
re q u ire s (a) three n ig h ts (b) five nights (c) seven
nights to re a d ju s t completely to your new ro u tin e .
17. If you cut your needed amount of sleep tonight by 50/£,
to complete the same amount of work tomorrow th a t you
o rd in a rily do, you w ill require (a) 40^» more energy
(b) 60/b more energy (c) lQQfo per energy.
18. In terms of n atio n al h a b its among c o u n trie s, the most
e f f i c i e n t sleep ers are (a) the Mexicans (b) the B ritis h
(c) the Germans.
19. Less than (a) 20/6 (b) 40>o (c) 6Q/o of married couples
sleep in single beds.
20. Business executives need (a) h alf as much sleep (b)
twice as much sleep (c) about o n e-th ird as much sleep
as the average manual la b o re r.
21. A short nap in the daytime has (a) no e ffe c t on one's
a b i l i t y to get to sleep at night (b) great e f f e c t on
one's a b i l i t y to get to sleep at night.
22. Nervous, ex c ita b le persons (a) need more sleep and get
le ss (b ) need le s s sleep and get more (c) need more
sleep and get more than p la c id , calm in d iv id u als.
23. The soundest sleep for the average adult i s (a) between
the t h ir d and fo u rth hours (b) between the f i f t h and
six th hours (c) between the second and th ir d hours.
124
24. (a) About bO/b (b) about 6 0 ft (c) about 70% of us Ameri
cans wear pajamas to bed.
2b. Hypnosis resembles sleep in th a t (a) the cortex of the
b rain is in h ib ited (b) the muscle system is completely
relaxed (c) the blood pressure is increased.
APPENDIX D
SUPPLEMENTAL TABLES
TABLE 18
TABLE OF R A W DA TA OF PAIRED SUBJECTS
Subjects W ho Listened to Chet Parker Subjects W ho Listened to Dr . Maloney
X
0 )
co
C C D
•H C T >
C D
f - l ’ — t
0 5 • — 1
C D O
>*o
c d
o n
<
a t
3
c o
tH
> e
a* c d M
< £ C l C D
O fth
+ - >
c
c d
e
C D C D
, , u N
U 4 c o O
UH O
< a, to
M - i
O + J
0 )
C D C D
C D H
U C D
an +j
C D C
a M
C D
( H
O + J
o ( f t
C O C D
H
2
(0 c
V C O
X
C D
C O
c a )
• h an
C D
H 1 — 1
C O 'H
C D o
> o
C D
O '
<
a
c o
■ H "H
> e
C L . C D U
< M C D
0Q,H
+ j
c
C D
s
C D C D
O f - l
U J (0 o
o r 1 o
< a. co
o 4 - >
< 0
C D C D
C D H
f - l C D
on+J
C D C
Q M
C D
O -fJ
U C O
C O C D
H
S
c o C
C C O
Male Fre sh. 18 1.63 3 HI* 21 M a le Fresh. 18 1.56 3 I 23
M a le Fresh. 18 1.67 4 I
22 Male Fre sh. 18 1.73 4 I 22
Male Fresh. 18 2.08 4 HI 22 Male Fre sh. 18 2.00 4 I 21
Male Fresh. 18 2.27 4 I 19 M a le Fre sh. 18 2.19 4 I 23
Male Fresh. 18 2.47 5 I
21 Male Fre sh. 18 2.33 5 HI 25
M a le Fre sh. 18 2.60 5 1 22 Male Fre sh. 18 2.55 5 I 22
Male Fresh. 18 2.60 5 I 20 Male Fre sh. 18 2.70 5 HI 21
Male Fresh. 18 2.69 6 I 24 M a le Fre sh. 18 2.80 6 HI 24
Male Fresh. 18 3.38 7 I 24 Male Fre sh. 18 3.29 7 HI 23
M a le Fresh. 18 3.66 9 HI
22
M a le Fre sh. 18 3.75 9 HI 23
M a le Soph. 19
1.82
3 I 18 Male Soph. 19 1.80 3 I 21
Male Soph. 19 2.00 4 I 21 Male Soph. 19 2.15 4 I 22
M a le Soph. 19 2.20 4 I 20 Male Soph. 19 2.31 4 I 23
M a le Soph. 19 2.39 4 I 20 Male Soph. 19 2.50 4 I 20
M a le Soph. 19 2.53 5 HI 23 M a le Soph. 19 2.43 5 I 20
M a le Soph. 19 2.95 7 HI 23 Male Soph. 19 2.93 7 I 22
Male Soph. 19
3.12 7 I
21 Male Soph. 19 3.10 7 HI 22
M a le Soph. 19 3.29 7 HI 20 M a le Soph. 19
3.22
7 HI 23
Male Soph. 19 3.33 7 I
22
Male Soph. 19 3.30 7 I 21
*HI re p rese n ts highly in te re s tin g ; I rep rese n ts in te r e s tin g .
i— 1
fO
o
TABLE 18—C ontinued
Subjects W ho Listened to Chet Parker Subjects W ho Listened to Dr. Maloney
X
Q J
C O
C 0
•h cn
0
f-t -H
0 ’ — 1
0 O
> o
0/
C T >
<
m
p
c o
•H -H
> E
O il 0) f-i
< C f H 0
O M - 1
+j
c
0
E
a > at
o fn
lu nj o
O r -H O
< O h C O
4-i
o + - >
in
0 (D
0 f H
fn 0
c r > + ->
0 c
Q n
0
f H
O 4 ->
o
U 7 0
H
3
0) C
c C o
X
0
C O
C 0
•H C P
0
f H ■ — <
(0 '— 1
0 O
> o
0
C P
<
1 0
p
c o
> E
0 4
< f H 0
O cu H
H - >
c
0
E
0 0
W o o
O — • o
<. C 4 C O
4-i
° K
0 0
0 f H
fH 0
cn+j
0 C
p l - l
0
f H
O H - >
o in
C O 0
f-*
3
(fl c
c C o
Male Soph. 19 2.80 8 I
22 M a le Soph. 19 3.00 8 HI
21
Fem. Fresh. 18 1.81 3 I 20 Fem. Fre sh. 18 1.81 3 I
21
Fern. Fresh. 18 2.1b 4 I
21
Fem. Fresh. 18 2.00 4 I
22
Fem. Fre sh. 18 2.36 5 I 22 Fem. Fresh. 18 2.29 5 HI 19
Fem. Fresh. 18 2.49 5 I
21
Fem. Fre sh. 18 2.38 5 HI 20
Fem. Fresh. 18 2.65 5 HI 22
Fem. Fre sh. 18 2.44 5 I
22
Fem. Fresh. 18 2.13 6 I
22
Fem. Fre sh. 18 2.00 6 I
22
Fem. Fre sh. 18 2.56 6 I
22
Fem. Fresh. 18 2.71 6 I
21
Fem. Fresh. 18 3.06 8 I
21
Fem. Fresh. 18 3.22 8 I 23
Fem. Fre sh. 18 3.35 8 I
22
Fem. Fre sh. 18 3.28 8 HI 22
Fem. Fresh. 18 3.19 9 I 22 Fem. Fresh. 18 3.29 9 HI 24
Fem. Soph. 19 2.00 4 I 20 Fem. Soph. 19 2.10 4 HI 20
Fem. Soph. 19 2.19 5 HI 21
Fem. Soph. 19 2.07 5 I 19
Fem. Soph. 19 2.42 5 HI 21
Fem. Soph. 19 2.47 5 HI 25
Fem. Soph. 19 2.43 6 I 19 Fem. Soph. 19 2.36 6 HI 23
Fem. Soph. 19 2.59 6 HI 22
Fem. Soph. 19
2.62 6 HI 20
Fem. Soph. 19 2.70 6 HI 22
Fem. Soph. 19 2.71 6 HI 22
Fem. Soph. 19 2.85 6 I 21
Fem. Soph. 19 2.76 6 HI
22
Fem. Soph. 19 3.10 7
, I
23 Fem. Soph. 19 3.25 7 HI 23
Fem. Soph. . 19 3.60 9 I 22
Fem. Soph. 19 3.50 9 HI 23
Fem. Soph. 19 4.00
9 I 23 Fem. Soph. 19 4.00 9 HI
22
128
TABLE 19
COMPARISON OF TEST SCORES A ND LEVEL OF INTEREST
RELATED TO SUBJB3TS SEATED IN CHAIRS 2, 3,
AND 5 AS DESIGNATED IN FIGURE 1
Chair 2 Chair 3 Chair b
2. 22
I b. 21
I 9. 24 I
18. 20 HI 19.
22
I 10.
22 HI
21. 20 I 2b. 22 HI 22. 21 I
40. 23 I 38. 23 I 36..
22 HI
1 . 21
HI 7. 20 I 3. 22 HI
24. 21 I lb. 23 HI 14. 20 I
27. 22 I 33. 21 HI 29.
22
I
3b. 22 HI 34. 19 I 37. 21 I
3. 21
I b. 2b HI 8. 24 HI
18. 23 HI 14. 20 I 12. 22
I
24. 20 HI 27. 21 I 21. 21 I
32.
19 I 29. 22 HI 30. 24 HI
4. 23 I 7. 21 HI 11.
21 I
19.
21 I 13. 23 I 22. 22 I
2b. 22 I 31. 20 HI 23. 19 HI
38. 23 HI 3b. 20 HI 40. 22 HI
343 343 349
NOTE: Under each heading, the f i r s t column rep re
sents the order on the Table of Raw Data, the second column
re p re se n ts the t e s t score, and the th ir d column re p o rts the
level of in te r e s t in the rad io ta lk . The scores are re
corded without c o rre c tio n for omissions. This c o rre c tio n
would not be s u f fic ie n t to indicate a trend of su p e rio rity
for any p a r tic u la r seating placement. C hairs 2, 3, and 5
represent the c h a irs at the two extreme p o sitio n s of c lo s e s t
and f a r th e s t away from the speaker and the cen ter p o sitio n .
Observation of t h i s data does not indicate an advantage for
any p o sitio n in determining r e s u l t s of the study.
129
TABLE 20
Chet Parker
F i r s t Week of Testing
No. on Raw
Score
Degree of
Data Table In te re s t
2 22
I
4 19 I
5 21
I
6 22
I
9 24 I
10 22
HI
11 18 I
17 21
I
18 20 HI
19 22
I
20 22
I
21
20 I
22 21
I
2b 22
HI
30 22
I
32 21
HI
36 22
HI
38 23 I
40 23
407
I
Third Week of Testing
1 21
HI
3 22
HI
7 20 I
12 21
I
13 20
I
14 20
I
lb 23
HI
16 23
HI
'24 21
I
26 22
I
27 22
I
28 21
I
29 22
I
31
20
I
130'
TABLE 20—C on tin u ed
Chet Parker
Third Week of T esting
No. on Raw
Data Table
Sc ore
Degree of
In tere st
33 21 HI
34 19 I
35 22 HI
37 21 I
39
22 I
403
Dr. Maloney
Second Week of T esting
2 22
I
3 21
I
5 25 HI
6 22
I
8 24 HI
10 23 HI
12 22
I
14 20 I
15 20 I
18 23 HI
21 21
I
24 20 HI
27 21
I
28 23 I
29
22
HI
30 24 HI
32
19 I
33 25 HI
397
Fourth Week of Testing
1 23 I
4 23 I
7 21
HI
9 23 HI
11 21
I
TABLE 20—C ontinued
131
Dr. Maloney
Fourth Week of Testing
No. on Haw
Score
Degree of
Data Table In tere st
13 23 I
17 22 HI
19 21 I
20 21 . HI
22 22
I
23 19 HI
25 22
I
26 22 I
31 20 HI
34 23 HI
35 20 HI
37 22 HI
38 23 HI
40 22 HI
413

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

An Experimental Study Of The Influence Of Subliminal Cue Words On Audience Responses To A Filmed Speaker'S Sincerity, Effectiveness, And Subject Matter

PDF

An Experimental Study Of The Retention And Comprehension Of Poetry Resulting From Silent Reading And From Oral Interpretation

PDF

An Experimental Application Of 'Cloze' Procedure As A Diagnostic Test Of Listening Comprehension Among Foreign Students

PDF

Critical Study Of The Nominating Speeches At The Democratic And Republican National Conventions Of 1960

PDF

An Empirical Study Of The Behavioral Characteristics Of Sincere And Insincere Speakers

PDF

A Historical Study Of Speech Education At The University Of Southern California (1880 Through 1950)

PDF

An Empirical Study Of Self-Feedback During Speech Communication

PDF

Teaching Operative Dental Technics Via Television: An Experimental Inquiry

PDF

An Experimental Study Of The Effect Of Listener Feedback On Speaker Attitude

PDF

An Experimental Study Of Some Effects Of Time Compression Upon The Comprehension And Retention Of A Visually Augmented Televised Speech

PDF

An Analytical-Historical Study Of The Factors Contributing To The Successof Mark Twain As An Oral Interpreter

PDF

A Survey And Analysis Of Current Attitudes Toward Censorship Of The Legitimate Theatre In The United States

PDF

An Experimental Study Of Intelligibility Thresholds Of Hypacusic Adults Using Forced-Choice And Yes-No Techniques

PDF

Content Analysis Of Fifty-Three High School Basic Speech Texts, 1899-1969

PDF

An Experimental Comparison Of Spoken Communication Developed Individually and Interindividually

PDF

A Historical Study Of The Belasco Theatre In Los Angeles And The Forces That Shaped Its History: 1927-1933

PDF

An Experimental Study Of The Accuracy Of Experienced And Inexperienced Speakers In Identifying Audience Behavior As Indicative Of Feelings Of Agreement, Indecision, Or Disagreement

PDF

A Historical Study Of The Characteristics Of Acting During The Restoration Period In England (1660-1710)

PDF

An experimental study of audience responses to a play rehearsed with and without a play analysis by a professional psychoanalyst

PDF

An Experimental Study Of The Ability Of Lay Judges To Distinguish Betweentypescripts Of Individual Idea Development And Group Idea Development

Asset Metadata

Creator Livingston, Harold Maurice (author)

Core Title An Experimental Study Of Effects Of Interest And Authority Upon Understanding Of Broadcast Information

Degree Doctor of Philosophy

Degree Program Speech

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag mass communications,OAI-PMH Harvest,Speech Communication

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Harwood, Kenneth A. (committee chair), Dickens, Milton (committee member), McCoard, William B. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-237474

Unique identifier UC11358088

Identifier 6106297.pdf (filename),usctheses-c18-237474 (legacy record id)

Legacy Identifier 6106297.pdf

Dmrecord 237474

Document Type Dissertation

Rights Livingston, Harold Maurice

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA