The effect of physical exercise on cognitive and psychological functioning in community aged

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content THE EFFECT OF PHYSICAL EXERCISE ON COGNITIVE AND PSYCHOLOGICAL FUNCTIONING IN COMMUNITY AGED by Charles Fiske Emery A Dissertation Presented to the FACULTY OF 'I'HE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Psychology) December 1985 UNIVERSITY OF SOUTHERN CAUFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 90089 This dissertation, written by ..................... ~~+~.Sl .. f.~~~~.;~:n' ...................... . under the direction of h.is ....... Dissertation Committee, and approved by all its members, has been presented to and accepted by The Graduate School, in partial fulfillment of re quirements for the degree of DOCTOR OF PHILOSOPHY .)t~,c;/1 d~~ JY". r(j~· ~j'Graduate Studies Date ..... R~F.9.l:>.~~ .. .t~.\ ... ~~?.?. ....... DISSERTATION COMMITTEE ·····~---~'S;·················c~,;;;;p;;;;; }[7)/} ' '""<'l/-- j l.!:.~ .. ~~:~f"3~( .. :f:~(~ ................................... . <-J(a{;t~>l' 1 {wvAJei 1 ........................... !!.r.~~ ............... i,; ..................................... . ~-=L::;:·•to•:;, ~n ; / '"': ...... ,:::~ ~i' · ..... ~ \'h.D. rs '85 t.53 DEDICATION To aerobics fans of all ages. ii ACKNOWLEDGEMENTS I am grateful to a number of people for their assis tance in completing this dissertation. First, I would like to thank the AARP Andrus Foundation for providing the necessary funds. I also wish to thank the staff of the Agape Senior Center, in particular Judith Morrison and Sandra Howard, for their assistance. I especially wish to thank the older adults at the Agape Center who participated in the project and who generously gave of their time during the lengthy testing procedure. They were a fascinating and pleasurable group to work with. I am grateful to the instructors for the classes, Linda Cassidy, Valentine Villa, and Claudia Graham, without whom the project would not have been possible. In addi tion, I wish to thank John Maguire, Candace Stacey, Melinda Garcia, Alison Smith, and Sheilah Sulliger whose help at various stages of the project was invaluable. Furthermore, I greatly appreciate the aid of the Andrus Volunteers of the Andrus Gerontology Center who assisted with piloting of measures. I am indebted to my committee members Scott Fraser, Steve Lopez, Al Marston, and Robert Wiswell all of whom contributed both their time and thoughtful suggestions to iii this project. Most important, I wish to thank my chair, Margaret Gatz, who has been enthusiastic about this pro ject from its genesis and has encouraged me not only in this study but also in the pursuit of meaningful research and fulfilling academic and personal experiences through out my graduate training. Her intellectual rigour and curiosity, and her ability to manage a multitude of people and projects with finesse, have been a motivating force for me during the past three years. Finally, I wish to thank my family for their encour agement, especially my mother for her inspiration and, above all, my wife Edith Pattou Emery who, after suffering through many long and lonely nights and weekends, still re mains my greatest love and support. iv TABLE OF CONTENTS ii DEDICATION • • • • ACKNOWLEDGEMENTS • CHAPTER . . . . . . . . iii I. II. III. IV. INTRODUCTION • • • • • • • • • • • • • • • LITERATURE REVIEW METHODS • • • • • A. Subjects B. Procedures C. Measures • RESULTS • • • . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . 1 6 • 20 22 • 25 • 30 • 39 81 96 REFERENCES • • • • • APPENDIX: MEASURES • • 1 0 1 v LIST OF TABLES Table 1. Design of the Study •••• • • • • • • • 2 1 2. Frequencies and Percentages for Demographic Measures • • • • • • . . . 3. Measures ••••••• . . . . . . . 4. Intercorrelations Among Physiological Measures at Time 1 • • • • • • • • • 24 31 41 5. Intercorrelations Among Fluid Intelligence Measures at Time 1 • • • • • • • • • • • • 44 6. Intercorrelations Among Psychological 7. Measures at Time 1 • • • • • Pre-Test Mean Scores and Standard Deviations • • • • • • • • • • • 8. Pre and Post-Test Mean Scores and . . . . . 45 47 Standard Deviations • • • • • • • • • • 53 9. Pre and Post-Test Mean Scores and Standard Deviations for Females • • • • 56 10. Mean Scores and Standard Deviations for Time 3 Follow-Up • • • • • • • • • • • 71 vi CHAPTER I INTRODUCTION Physical exercise in the form of aerobic exercise has been widely publicized and advertised in the media in re cent years. Books describing different exercise programs are flourishing and have become national best-sellers; increasing numbers of televised exercise programs are avail able to the general public; and in Los Angeles one radio station even conducts a morning aerobic workout for its listeners. Exercise has become a veritable panacea for our society and there is a great deal of social reinforcement for participating in regular physical exercise. At the same time, physical exercise is frequently included by physicians as part of prescribed health regimens for older adults, based on the notion that it will benefit the physi cal health of participants and perhaps improve their psy chological well-being. This raises numerous questions about the nature of the benefits of exercise. Particularly among older adults this is an interesting issue which has received insufficient attention in the research literature. While there is a large amount of literature describ ing the general beneficial effect of physical activity on the physical health of older individuals, there is rela- 1 tively little discussion of the effect of exercise on the mental health of these individuals. There are self-re ports by older people suggesting that they feel better, sleep better, and feel less angry as a result of physical exercise (Harris & Frankel, 1977). Indeed, it has been thought that both behavioral and physiological factors resulting from regular exercise contribute to mental in tegrity (Powell, 1975; Barry, Steinmetz, Page & Rodahl, 1966), and that alleviation of certain physiological condi tions improves overall mental performance, particularly cognitive performance (Felkins & Sime, 1981). However, past research examining the psychological and cognitive changes resulting from physical activity have been limited by design deficiencies. Those studies with adequate con trols have examined only institutionalized geriatric popu lations, thereby limiting the applicability of the results. Research has shown that physical exercise improves cardiac functioning and keeps blood pressure low in older individuals (Harris & Frankel, 1977). However, there have been few studies which indicate psychological variables upon which physical exercise has a beneficial effect. A key feature of this project is that it examined, in a con trolled experiment, the effect of physical activity upon cognitive and mental health variables which are central to the well-being of older individuals. A second feature of this study was the use of ecolog- 2 ically valid measures. Previous studies have tended to look at changes in personality factors and in performance in crystallized intelligence, both of which remain rela tively stable with age (Stamford, Hambacher & Fallica, 1974; Buccola & Stone, 1975}. This project examined cogni tive faculties which are known to decline with age, called fluid intelligence (Horn & Donaldson, 1976}, as well as psychological well-being. Fluid intelligence involves the ability to concentrate, to attend, and to reason logically. Psychological well-being was defined in terms of depressive symptomatology, self-rated anxiety, self-rated happiness, locus of control, and sense of mastery. The personal con trol construct was of particular interest because it has been highlighted by a number of theorists defining positive psychosocial adaptation in older adults (e.g., Rodin, 1980; Kuypers & Bengtson, 1973). In order to assess the personal control dimension this study included both a generalized measure of locus of control (Rotter, 1966) and a locus of control scale specific to health (Lau & Ware, 1981}. A third feature that guided this study was generaliza bility of results. The past research in the area with the best designs and the most adequate controls was performed with institutionalized groups. This study examined a non institutionalized elderly population in order to insure the generalizability of the results to a broad aging popula tion. In other words, it would be expected that older 3 adults who participated in this project resemble those who participate in exercise programs in the community, at senior centers as well as at other locations. In short, it was predicted that elderly subjects, af ter a physical exercise program, would report improvement on a set of dependent variables selected to encompass physiological fitness (weight, flexibility, heart rate, blood pressure), cognitive functioning, and psychological well-being. It was hypothesized that these subjects, as a result of the exercise program, would demonstrate an in creased capacity for fluid intelligence and, of central concern, lowered depression, and enhanced sense of mastery, personal control, and self-efficacy. Thus, it was thought that both increased capacity for fluid intelligence and greater psychological well-being would be related to im provements on physiological indices. Physical exercise would stimulate blood circulation and improve cardiovascu lar functioning, and thereby cognitive capabilities would be revivified and sense of well-being would increase. Physical exercise was in the form of aerobic condi tioning classes. The experimental subjects were compared to two control groups, one involved in non-physical activ ity, the other placed on a waiting list. Physiological measures were taken of all three groups to insure that a physical training effect was established in the exercise group which, in turn, was expected to account for any 4 achieved relationship between physical and psychological change. A randomized experimental design was used to test the following hypotheses: that, compared to the control groups, participants in the exercise program would indi cate less depressive affect at the end of the exercise pro gram, that they would shift from a more external to a more internal locus of control on both generalized and health related measures of locus of control, and that they would report a greater sense of mastery over their lives, greater happiness, and lower levels of anxiety. It was also hy pothesized that, compared to the control groups, the exper imental group would improve on tasks involving fluid intel ligence and improve on the physiological measures. Final ly, it was hypothesized that the exercise group would be likely to endorse more highly various health behaviors as a result of having participated in the exercise program. .5 CHAPTER II LITERATURE REVIEW The sections below expand on the literature pertinent to this study. The first section reviews research about the psychological effects of exercise on younger popula tions; the second section discusses the literature regard ing physical exercise among older adults; the third sec tion describes the implications of prior research for the design of this study; and the latter four sections expli cate the specific dependent measures on which effects were predicted for the elderly in this study. Physical Activity and Younger Populations Studies of the effect of regular physical exercise on personality characteristics of younger populations have indicated that mood factors such as depression and anx iety decrease, while happiness increases (Carter, 1977; Folkins, 1976). Morgan and his associates (Morgan, Roberts, Brand & Feinerman, 1970) investigated the rela tionship between depression and selected variables in nor mal middle-aged males. Using the Zung Self-Rating Depres sion Scale, they found that depression was not significant ly related to age, height, weight, percent body fat, strength of grip, or physical work capacity. Contrary to 6 other results, they also found that a six-week program of exercise did not produce significant changes in depression. One problem with the latter finding is that the physical activity program was not of sufficient intensity or dura tion to evoke physiological change and, in fact, no fit ness effect was demonstrated in this study. Indeed, the standard length of exercise programs for such studies is at least twice the length of theirs. Of interest, however, was the Morgan et al. (1970) finding that depressive affect decreased within the subset of the experimental group that was depressed at the outset of the experiment. Their study also made special note of the self-report by experimental subjects that they "felt better" after the exercise session and that 85% of the sub jects volunteered to participate in future studies of this type. These researchers pointed out that there were no adequate psychometric tools, at that time, to assess ob jectively the state of feeling better. This lack of ob jective psychometric evidence for the broad claims of psy chological improvement plagues much of the research in this area (Felkins & Sime, 1981; Clark, Wade, Massey & Van Dyke, 1975). In general, research with a non-aged population has indicated positive effects of physical activity on mood, self-image, and happiness. However, the flaws of the Morgan et al. (1970) study are characteristic of past 7 research in that there is lack of psychometric evidence for general positive conclusions, lack of a physical train ing effect, lack of randomized control group, and conse quently, inadequate manipulation of the independent var iable, physical exercise. Physical Activity and .Aging Studies of community-residing aged have provided equivocal evidence that exercise improves mood in this population. Tredway (1978) studied three groups involved in aerobics, calisthenics, and shuffle-board in an on going activity program at a retirement community. The shuffle-board group was the control group. This researcher found an acute and transient positive mood effect for physical activity, in both aerobics and calisthenics, when participants were tested immediately prior to activity and immediately following activity. She also found that de pression declined over the 15-week study period for calis thenics relative to aerobics and shuffle-board. However, two methodological problems are evident in this study. There was not random assignment of subjects to groups which may confound the physical activity variable and bias the results, and shuffle-board may not serve as an adequate control for a study of the effect of exercise. Buccola and Stone (1975) looked at the effect of phys ical exercise on the personality characteristics of aged 8 males using Cattell's 16 Personality Factors. These re searchers found that males aged 60-79 who were involved in a program of walking and jogging became significantly hap pier and more self-sufficient. They found no other signi ficant changes in personality factors. Two problems with this study are that no control group was used and only males were studied. Blumenthal and his associates (Blumenthal, Schocken, Needels & Hindle, 1981) found small and generally nonsig nificant changes in psychological variables among 24 sub jects who took part in an 11-week conditioning program. Two limitations of this study were that there was no con trol group and that the exercise sessions lasted only 30 minutes, three times per week, on a stationery bicycle er gometer. A more recent study (Perri & Templer, 1985) indicated that aerobic exercise among 23 older adults produced sig nificantly greater changes in locus of control attributions and in self-concept compared to a control group. However, it was not mentioned whether or not the exercise and con trol groups were randomly assigned. Thus, while all four of the aforementioned studies examined non-institutional ized aged, they lacked adequate control groups and/or sys tematic manipulation of the independent variable. Far better studies of the effect of physical activity on psychological variables have been done with institution- 9 alized geriatric patients. Three of these are described here. Clark et al. (1975) studied a group of 23 institu tionalized male and female geriatric mental patients. This study employed two control groups of which one was a social group given games to play. The experimental group took part in stretching exercises, rhythmical exercises, and dancing. All groups showed increased total daily activity level, but the exercise group increased the most. These researchers also found a large performance difference be tween males and females, indicating that males were more highly motivated to the task than females. Stamford et al. (1974) studied a group of institution alized subjects who walked on a treadmill daily for twelve weeks with sessions lasting from 6 minutes to 20 minutes. A control group walked on the treadmill each day for one minute and, thus, was exposed to a setting and social in teraction similar to that of the experimental group. The dependent variables were performance on the Draw-A-Person Test, on two subscales of the WAIS (Digit Span and Informa tion) as well as on a general hospital questionnaire. There were significant differences only on the WArS-Infor mation test and the questionnaire. Thus, subjects' in creased ability to retrieve information was noted as was the subjective observation that subjects became more so cially interactive as the study progressed. Powell (1974) used a two-control-group design in 10 another study of the effect of physical activity on insti tutionalized male and female geriatric patients. He found that subjects improved significantly on the Progressive Hatrices Test and Wechsler I-1emory Scale, after 12 weeks of exercise. The positive effect of exercise therapy was thought to be caused by a revitalization of mental abili ties already extant, but in a state of disuse. This was thought to result from physical activity since the social control group showed no improvement on these measures. The use of the social interaction control groups in each of these three studies increased the internal reli ability. At the same time, the use of institutionalized patients--and, in the case of Stamford et al. (1974), the use of only male psychiatric inpatients--limited the exter nal validity of these studies. This is particularly prob lematic in studies of cognitive changes resulting from physical activity. A final concern is that the choice of dependent measures may have limited the substantive validity. For instance, given that the WAIS-Information subtest may be biased by retest effects, improved per formance may not be indicative of improved cognitive abil ity. Implications for Design of Present Study It is clear from this review of past literature that the most extensive studies with the most convincing results 11 have been performed with institutionalized geriatric mental patients. However, because the majority of the elderly population is not institutionalized, it seems necessary to evaluate normal community-residing aging individuals in studying the effect of physical activity on cognitive and psychological variables. Hence, for this study subjects were recruited from a community center within the city of Los Angeles, and the study incorporated some of the design specifications of previous research, such as the social in teraction control group. Furthermore, objective psycho metric tests were used to determine the psychological ef fects of physical activity. Since there is no evidence to support the claim that global changes on personality tests follow from fitness training, it was necessary to focus on target variables that were expected to shift (Felkins & Sime, 1981). Specifically, measures of fluid intelligence, depression, locus of control, and self-efficacy were in cluded. Cognition This study examined the effect of physical activity on the types of cognitive functioning which have been called fluid intelligence. These include flexible problem solving and the capacity to integrate new information, as opposed to crystallized intelligence which includes recall of previously learned information. Fluid intelligence re- 12 fers to mental ability that is independent of acquired knowledge, experience, and learning, whereas crystallized intelligence refers to knowledge acquired through exper ience (Horn & Cattell, 1966). An example of a test of crystallized intelligence is the WAIS-Information subtest; an example of a test of fluid intelligence is the WArS Digit Symbol subtest. While crystallized intelligence im proves or increases during adulthood, fluid intelligence has been found to decline both with age (Horn & Donaldson, 1976) and with decreasing physical fitness (Elsayed, Ismail & Young, 1980). Fluid intelligence, therefore, was chosen as an outcome variable for this study because it is an area in which deficits are often present among older adults and where improvement is feasible. Moreover, be cause performance on tests of fluid intelligence has been linked to health and fitness, fluid intelligence is the aspect of cognitive functioning that is most likely to be affected by physical exercise. Whereas the intent of this study was to examine the effect of physical activity on cognitive functioning, cog nitive functioning--especially fluid intelligence--has also been shown to improve in relation to general positive at mosphere and expectations of good performance (Felkins & Sime, 1981; Elsayed et al., 1980). Thus, one function of the social activity control group was to facilitate posi tive expectations similar to those of the exercise group. 13 Then comparison of the two types of groups on the cognitive measures would control for the social factors affecting performance and would focus on the specific role of physi cal activity in changing cognitive capabilities of older individuals. Depression Depression represents a major health problem in modern society. Over 50% of psychiatric patients of all ages have depressive disorders. Among adults aged 65 and older, de pression is the most common functional disorder (Kay & Bergmann, 1980). At the same time, community-residing aged without diagnosed depression are nonetheless especially susceptible to transient dysphoric mood as indicated on self-report symptom check-lists (Blazer, 1980). Greist, Klein, Eischens, Faris, Gurman & Morgan (1979) studied the effect of physical activity on moderate depres sion in a group of patients between the ages of 18 and 30. These researchers compared the effects of running treat ment, time-limited psychotherapy, and time-unlimited psy chotherapy on the Symptom Checklist - 90 (SCL-90) depres sion cluster. They found that the running treatment was as effective in alleviating depressive symptoms and target complaints as either the time-limited or time-unlimited psychotherapy. The explanation for such an effect is multi-faceted. These researchers believe that their run- 14 ning patients developed sense of mastery over what they perceive to be a difficult skill, patience, capacity for change, generalization from changing one habit to changing others, and distraction from physical symptoms of depres sion. In a more recent study, McCann & Holmes (1984) found that aerobic exercise reduced levels of depressive symptom atology among a group of depressed female college under graduates. For their study, they randomly assigned sub jects either to the exercise group, to a relaxation group, or to a waiting list control group. The present study used a similar randomized control group design to examine depressive symptoms in an aging population. Although the sample was not selected with clinical depression as a criterion, depressive symptoms re flect the mood state of many participants, and prevalence studies among aging populations indicate that such symptoms are common complaints. The measure of depression used in this study, the Center for Epidemiological Studies depres sion scale (CES-D) has been standardized with normal com munity-residing aged (Himmelfarb & Murrell, 1983). In addition, two measures related to depression and self-efficacy were included in this study; a one-item mea sure from the CES-Anxiety scale, and the one-item avowed happiness measure from Bradburn (1969). 15 Locus of Control and Mastery The notion of locus of control is important in a de scription of the psychological well-being of older adults. An internal locus of control is defined in terms of the in dividual's feeling that outcomes in life are due to his or her own efforts. The opposite belief, namely that one's opportunities in life are ruled by chance or fate, is called external control. Kuypers and Bengtson (1973) postulate decreased per sonal control as central to understanding how societal ageism and individual life changes can have a negative ef fect on some persons. In pre- and post-retirement age in dividuals internal control has been found to be a key pre dictor of life satisfaction (Palmore & Luikart, 1972) and of adjustment (Kuypers, 1972). t..Vhile there is controversy about how locus of control may change with age (Brim, 1974), there is agreement on the value of fostering the older individual's sense of control and responsibility for his or her actions (Rodin, 1980). Furthermore, studies have indicated that establishing a contingent, i.e. controllable, environment leads to improved performance on memory tasks (Langer, Rodin, Beck, Weinman & Spitzer, 1979) and that there are relationships between internal locus of control and fluid intelligence (Lachman, Baltes, Nesselroade & Willis, 1982). Research with a non-aged population has indicated 16 the efficacy of exercise in bringing about a shift in per ceived locus of control. Duke, Johnson & Nowicki (1977) found that a group of 109 boys and girls aged 6 to 14 changed from a more external locus of control to a more in ternal locus of control as the result of an eight week sum mer sports fitness camp. The study found significant im provements in physical fitness variables also. Though it is impossible to make a direct causal inference between the physical exercise and increased feelings of control, the exercise is a plausible explanation for this shift. Perri and Templer (1985) also found shifts toward internality in their study of older adults engaged in aerobic exercise. Because locus of control is a generalized construct, it tends to be a fairly stable personal attribute. The difficulty in detecting change on measures of locus of con trol was addressed by employing multiple scales. Rotter's Internal-External Scale (Rotter, 1966) was used to measure generalized locus of control. In addition, a locus of con trol scale pertaining to health beliefs was used because scales specific to any particular domain of control are expected to be more sensitive to change (Rotter, 1975). In this case, Lau and Ware's Health Locus of Control Scale (Lau & Ware, 1981) was used to measure health-specific locus of control. Finally, the Pearlin Mastery Scale (Pearlin & Schooler, 1978) was used to measure mastery. Mastery is 17 defined as a sense of personal competence and control. It is thought to be a vital psychological resource which people draw upon to help themselves withstand threats posed by the environment. While past research has shown that older people generally score lower than young on mea sures of mastery (Pearlin & Schooler, 1978), there have been no studies published which attempt to intervene and increase sense of mastery in the elderly. Thus, this study evaluated improvement on participant's positive, effica cious self-concept using locus of control, mastery, and happiness. It was expected that a low score on the mastery scale would be correlated with a more external sense of control, and that shifts in both of these scores would oc cur as a result of physical exercise. Physiological Of equal importance for this study were the physiolog ical measures. Several studies in the past have not demon strated any kind of training effect of physical activity (Felkins & Sime, 1981). However, demonstration of a train ing effect is necessary to establish the hypothesis that physical activity rather than other variables causes the change in psychological and cognitive variables. In past studies of older adults, researchers have characterized and assessed physical change in this popula tion by using a variety of physiological measures. These 18 measures include resting systolic and diastolic blood pres sure, stress testing on a bicycle ergometer, weight, height, elevated systolic and diastolic blood pressure, skinfold measures, grip strength, heart rate, balance, total daily activity level, flexibility, power, and muscu lar endurance (Blumenthal et al., 1982; Clark et al., 1975; Morgan et al., 1970; Barry et al., 1966). The pres ent study incorporated several of these measures to assess change in participants' level of physical fitness. In par ticular, it was thought that resting and elevated heart rate measures would be valuable indicators of physiological change (Clark et al., 1975). In addition to the physiological measures, a one item self-rating of health was included. Furthermore, it was thought important to measure actions taken by individ uals to promote their health. Thus, subjects were asked to indicate levels of alcohol consumption, of smoking, and of physical exercise, as well as the degree to which they felt that their diet was balanced and their sleep was adequate. 19 CHAPTER III METHODS The experiment used a three-group pretest-posttest de sign. Subjects were randomly assigned to the experimental group or one of two control groups. The experimental group (E) was pretested, took part in 12 weeks of a physical ex ercise class, and was re-tested. The first control group (C1) was pretested, took part in social activity for 12 weeks, and was re-tested. The second control group (C2) was tested, was on a waiting list for 12 weeks, and was re tested. After the 12-week posttest, both control groups took part in the exercise program for 12 weeks, and all continuing participants were posttested a second time at 24 weeks. Table 1 summarizes this design. Group C1 controlled for the effect of the socializa tion aspect of exercise classes on the psychological state of the experimental subjects, while Group C2 controlled for the effect of history, maturation, instrumentation, and statistical regression. Use of a randomized experimental design maximized internal validity. Thus, while subjects may have varied in their level of activity outside of the experimental setting, randomization of subjects to condi tions controlled for the effect of this variable on the 20 Table 1 Design of the Study Pre- Post- Post- Test Test Test Group (T1) Condition {T2) Condition (T3) Experimental(E) X Exercise X Exercise X Control 1 (C1) X Social X Exercise X Control 2 (C2) X Waiting X Exercise X 21 test results. Subjects Subjects for the experiment were recruited from the Agape Senior Center in downtown Los Angeles. The Agape Center is affiliated with the Angelus Plaza, an innovative apartment complex developed jointly by federal, local, and private agencies, now housing more than 1400 predominantly low-income aged. The senior center serves residents of the apartment complex as well as non-residents living in the surrounding downtown community. In order to rationalize the program without revealing the experimental hypotheses, potential subjects were told that this was a 6-month program of both physical and non physical activities; three months of one kind of activity would be followed by three months of the other kind of activity. A number of recruitment strategies were implemented. First, an advertisement was placed in the monthly newslet ter, which is delivered to the door of every resident and is sent to every non-resident affiliated with the Center. This advertisement was followed by a 'Continental Break fast' at which the program was explained. Recruitment was conducted at monthly meetings of three other groups at the center, at a raffle during the senior lunch program, and at an afternoon dance (the 'Senior Prom'). Flyers and 22 announcements were posted and a recruitment table was set up during meals at the center. Furthermore, three other senior centers in the downtown area were contacted and notified of the program at the Agape Center. Flyers were sent to them and they agreed to notify the seniors at their centers. The procedures netted 57 persons who were interested in the program. Attrition was at about the level expected; when individuals were contacted by phone to arrange for the initial 'registration' interview and pretesting, 9 individ uals were unable to participate for various reasons. Thus,. participants in the study comprised 48 individuals, 40 females and 8 males, ranging in age from 61 to 86, with a mean age of 71.6. Participants were 46% Anglo, 23% Black, 25% Hispanic, 6% other ethnic groups, and they were pre dominantly low and moderate income; only one person lived outside the Angelus Plaza housing complex. More complete demographic information about subjects is listed in Table 2. Subjects were randomly assigned to the three condi tions, with 15 assigned to the exercise group, 15 assigned to non-physical activity, and 18 assigned to the waiting list. The assignment of a proportionately greater number to the waiting list was a precaution against anticipated drop-out from this group. Of the 18 individuals in this 23 Table 2 Frequencies and Percentages for Demographic Measures (N=48) Age N % - 60-69: 14 32 70-79: 25 57 80-89: 5 11 x=71.6 years Sex N % Male: 8 17 Female: 40 83 Marital Status N % Married: 9 20 Divorced: 5 11 Separated: 2 4 Widowed: 25 56 Single: 4 9 (never married) Ethnicity N % Caucasian: 22 46 Black: 11 23 Hispanic: 12 25 Asian: 2 4 American Indian: 1 2 Education N % 0-6 years: 7 16 7-11 years: 12 26 High school graduate: 17 38 Some education beyond high school: 9 20 Household Income N % $0-3,999: $4,000-5,999: $6,000-9,999: $10,000- Profession Manager of business, administrative, ser- 7 19 8 4 N vice professions: 10 Technicians, clerical, sales: 13 Semi-skilled worker or laborer: 13 Housewife: 9 19 50 21 10 % 22 29 29 20 Note: Ns may total less than 48 where there were missing data. 24 C2 group, 11 were part of the initial randomization pro cess, and 7 were late recruits, pretested during the first week of the program. Subsequent correlated t-tests on all of the dependent variables indicated that there were no significant differences between those initially assigned to the waiting list group and those who augmented the sample. Thus participants are considered together in all tests of the effects of the intervention. Procedures Subjects consented to take part in activities of both a physical and a non-physical nature and they agreed to respond to written questionnaires and to have physiological measures taken of them. Subjects were not told the hypoth eses of the experiment until after the experiment was com pleted. At the outset of the study all subjects were told that the experimenter was interested in their reaction to dif ferent activities at the senior center. At that time the waiting list control group was told that there were not enough instructors for the entire group but that there would be more instructors in the near future and that the experimenter still would like them to take part in the study. At the end of 12 weeks this waiting-list group was invited to take part in the physical exercise class. At that time the social activity group was also invited to 25 change to physical activity and the exercise group was given the option of either remaining in exercise or ending their participation in the activity program altogether. Exercise Program The exercise program consisted of warm-up and simple aerobic exercises. The intensity of the work-out in creased over a 12-week period through increments of number and muscular stress of exercises. Exercise classes were taught by an instructor who had had experience working with older adults in exercise groups. The instructor was em ployed for the purpose of running the exercise classes and was given the same rationale for the study as was given the subjects. This was to control for any differential effects of expectation within and between the two classes. Exer cise classes were held three times per week for one hour each, over a 12-week time period. An additional instructor was employed during the second 12-week period to teach the second exercise class comprised of people from the control groups. At the beginning of the program, participants were en couraged to judge for themselves their own level of physi cal exertion and to go at their own pace throughout the program. At the same time, participants were instructed to work themselves a little harder each session. This was facilitated by the planning of the exercise classes: while the basic exercise routine remained the same over the 26 course of 12 weeks, the number of repetitions of each exer cise was increased, thereby assuring increased muscular and cardiovascular endurance. Participants' blood pressure and heart rate were mea sured before the program began and participants with high blood pressure were encouraged to consult with a physician before beginning the exercise program. However, for a num ber of reasons no attempt was made to insure that the phys ician had been contacted. First, if anyone other than the participant had contacted the physician, this would have been intrusive. On the other hand, requiring participants to bring a note from their physician might have been in fantilizing or might have created a financial barrier for some individuals. Instead, several precautions were taken to prevent or minimize injury during the program. Classes were held in a community room directly above a health center which is affiliated with the senior center. The physician and nurse practitioners at the health center were informed of the ex ercise program and were warned of the possible need for their services in case of an accident. Furthermore, the exercise program was designed so that the instructor and the exercise participants monitored the level of physical exertion throughout each class. At the beginning of each exercise session, partici pants checked their heart rate and reported it to the 27 class. Each session then began with 5-7 minutes of rapid walking around the exercise room and onto an adjoining patio area. At the end of the walking, participants again checked their pulse and reported it to the class. Next, participants were guided through a series of systematic muscle stretching and strengthening exercises while seated in an armless folding chair. The exercises commenced with stretching the upper torso, back, neck, arms and legs. This was followed by arm circling to strengthen the arms, andlegkicking to strengthen the legs. Participants re mained seated while they then took part in exercises to strengthen and stretch the torso. This was followed by hand exercises with the arms extended in front of the body, and by heel and toe lifts to strengthen the legs. Participants were then led in an exercise in which they stood up and sat down repeatedly, finishing with the standing position. From there participants were led in side bends and knee-bends, as well as in a variety of leg strengthening exercises, including deep knee bends and leg lifts--to the front, to the back, and to the side--from a standing position. The exercise class always concluded with a dance routine accompanied by a popular song. Music accompanied the entire exercise session and included march ing songs as well as 'Swing Era' vocal music. At the end of class, heart rate was again measured by all partici pants. 28 The two references most useful in planning the exer cise program were Sager (1984) and Frankel (1977). The ex ercises took into account the types of bone and muscle changes characteristic of normal aging, e.g., osteoporosis. Thus, stretching exercises were emphasized rather than ex ercises that would strain the body, such as jumping or jog ging. The major aerobic component of the classes was the rapid walking and the dancing. However, throughout the hour, breathing was emphasized--both taking in oxygen, and forcefully expelling carbon dioxide from the lungs. This was thought to contribute to the overall aerobic condi tioning of participants. Social Activity Program The social activity program was comprised of various games and art projects. The activity sessions were led by an individual who was experienced in working with and in developing programs for older adults. This leader, also, was blind to the hypotheses of the study and was employed solely for the purpose of directing the activity sessions. Among the activities of this group, playing poker was the most popular. Other activities included making ceramic cups or pots, playing "Name That Tune" in which partici pants competed to identify a group of current and past musical hits, political discussion groups, and viewing of humorous film strips. 29 Measures The measures used in this study are of three general categories: physiological/health, cognitive, and psycho logical well-being. All measures were taken at the times indicated in Table 1. A complete list of measures is pro vided in Table 3. The psychological well-being scales and health and physical activity measures were included in a self-administered questionnaire. A copy of it is provided in the Appendix. 1) Physiological/Health Each participant was measured for resting heart rate and blood pressure, weight, arm strength, flexibility, balance, and heart rate response during a modified step test (Minister of State: Fitness and Amateur Sport (Canada), 1979). Heart rate response was measured in three segments of the step-test. Those individuals who did not have extremely elevated resting blood pressure (cutoff of 170/90) or resting heart rate (cutoff of 100) were asked to walk up and down two steps for three periods of three minutes each. The steps were constructed of ply wood with a 7" rise and were placed against a wall. Par ticipants stepped at three tempos: 50 steps/minute, 60 steps/minute, and 80 steps/minute. The tempo was main tained by a Wittner metronome. A 60-second break was al lowed between each segment, though this was difficult to enforce as many participants requested a longer break. 30 Table 3 Measures 1) Physiological/Health a) Resting and elevated heart rate b) Resting and final blood pressure c) Weight d) Sitting flexibility e) Standing flexibility f) Arm strength g) Balance h) Self-rated health i) Health-related behaviors 2) Cognitive Functioning a) Raven's Progressive Matrices Test b) Wechsler Adult Intelligence Scale (WAIS-R) - Digit Span subtest c) WAIS-R - Digit Symbol subtest d) Copy numbers test e) Copy words test 3) Psychological Well-Being a) CBS-Depression Scale b) CES-Anxiety measure c) Rotter Internal-External Scale d) Health Locus of Control Scale e) Pearlin Mastery Scale f) Bradburn self-rated happiness g) Pleasurable activities scale 31 Because there were no standard measures of strength and flexibility available for use, a battery of measures was devised using a physical fitness test manual (American Alliance for Health, Physical Education and Recreation, 1979) and with the consultation of an exercise physiolo gist, Dr. Robert Wiswell, whose research and prior consul tation includes work in gerontology. Arm strength was measured by the length of time that subjects held a three pound weight with each arm extended directly out to the side. When an individual's arm had lowered six inches or more, time was stopped whether or not the person was fatigued. Flexibility was measured in two ways. First, participants sat on the floor (or on a table), placed their feet flat against a wooden crate with a ruler at tached to it and, keeping their legs straight, reached for ward as far as possible, touching the ruler with both hands. Second, participants stood against a wall and mea surements were taken of the distance that they could reach down on each side, keeping their hands against the wall and, again, keeping their legs straight. Balance was mea sured by the length of time that participants were able to stand on each foot alone, with no other support. Partici pants were weighed, to the nearest tenth of a kilogram, on a HOMS full-capacity beam scale. The health and activity measures were included as part of the questionnaire that subjects filled out independent 32 of the examiner, after the physiological and cognitive testing was completed. This questionnaire is included in the Appendix. Self-rated health was assessed by asking subjects to rate their health as excellent, good, fair or poor (commonly included in self-report scales of physical health, e.g., the OARS, Duke University, 1978). Subjects were also asked to rate on a four-point scale the extent of their alcohol consumption, of smoking, and of physical exercise, as well as the degree to which they believed their sleep was adequate and their diet balanced (Moos, Cronkite, Billings & Finney, 1983). In a later section of the questionnaire subjects were asked to indicate the ex tent of their usual participation in mildly strenuous ac tivity, very strenuous activity, and activity in general (Moos et al.). The purpose of these measures was to as sess perceived level of involvement in activities such as housework that were not described as physical exercise. 2) Cognitive Functioning Cognitive measures were all intended to evaluate fluid intelligence. They included Raven's Progressive Matrices Test as well as two subtests of the Wechsler Adult Intelligence Scale (WAIS-R}--Digit Span and Digit Symbol--and two tests of writing speed (from the first NIMH longitudinal study of normal aging, described by Birren & Botwinick, 1959}. Raven's test has been normed on aged individuals and has been used in past research as 33 a measure of fluid intelligence (Powell, 1974). It in volves completing 36 separate designs each of which has a piece missing. For each design the subject is provided with 6 different alternatives from which to choose and is told to select the piece which best fits into the design. Thus, it evaluates problem-solving capacity independent of acquired knowledge and learning. The two WAIS subtests were Digit Span and Digit Symbol which were intended to measure fluid intelligence rather than crystallized intelligence. The Digit Span subtest measures the subject's ability to concentrate and to re call a list of digits, i.e., information that is not ver bally meaningful. It is a measure of primary memory. The Digit Symbol subtest and the two writing tests evaluated psychomotor performance, dexterity and mental speed. The first writing test required that subjects copy a page of numbers as quickly as possible. A subject's score was the length of time taken to copy the entire page. The second writing test required that subjects copy a page of English words as rapidly as possible, and the score was the number of words copied in two minutes. For the Digit Symbol test subjects were first presented with the digits 1-9, each paired with a different symbol. Below that were four rows of digits printed in random order, each paired with empty boxes. Subjects were required to write the symbol paired with each of the digits in the box, as rapidly as possible. 34 A subject's score was the number of symbols written in 90 seconds. 3) Psychological Well-Being The psychological measures were the depression scale from the NIMH Center for Epidemiological Studies (CES-D), a one-item rating of anxiety from the CES anxiety scale, Rotter's (1966) locus of control inventory, Lau and Ware's (1981) health locus of control inventory, the Pearlin Mastery Scale (Pearlin & Schooler, 1978), and a standard one-item self-rating of avowed happiness (Bradburn, 1969). Included with these were a list of activities, from danc ing to card playing, which subjects were asked to rate ac cording to how much they enjoyed the activity. This ac tivity list was derived from a standard Pleasant Events Schedule (Lewinsohn, Munoz, Youngren & Zeiss, 1978; adapted for older adults by Gallagher & Thompson, 1981). The CES depression scale {Radloff, 1977) was selected because it is widely used in epidemiological and evalua tion research to assess depressive affect with adults of all ages. Because most self-rated depression inventories have a high loading of somatic complaints, symptom levels are higher among elde~ly populations (Blazer, 1980). The CES-D is the only depression scale which is exceptional in this respect and, thus, it was thought to constitute a more accurate measure of depression for this study. Another measure of interest for this study was sub- 35 jects' self-rating of anxiety. While depressive symptom atology encompasses most of the negative moods commonly ex perienced, the CES-D does not inquire about feelings of tension, which are often another way of expressing a neg ative sense of psychological well-being. For this dimen sion a single item was selected from the CES anxiety scale and added to the end of the depression scale. This item reads, "During the past week I felt anxious or tense", and subjects were asked to choose from four possible responses "Rarely or none of the time, some of the time, occasional ly, or most of the time". The Rotter (1966) locus of control scale was selected because it is currently the most common measure of locus of control in adults and aged (Gatz, Siegler, George & Tyler, 1985; Kuypers, 1972; Rotter, 1975}. This scale includes 23 forced-choice items scored for externality. The follow ing is a typical item. "A. Becoming a success is a matter of hard work; luck has little or nothing to do with it." vs. "B. Getting a job depends mainly on being in the right place at the right time." Lau and Ware's (1981} health locus of control inven tory was included to complement the generalized Rotter locus of control scale and to assess change in ideology about control over health among subjects. This scale is composed of 27 statements regarding control of health out comes, each rated on a 7-point likert scale. The total 36 health locus of control scale is comprised of four sub scales: belief in provider control (e.g. physician), belief in self-control, belief in chance events, and belief in general health threats. The seven Pearlin mastery items (Pearlin & Schooler, 1978} provided a second measure of generalized perceived control. This scale consists of seven statements ranging from feelings of mastery to feelings of low personal effec tance, which respondents are asked to rate on a four-point scale ranging from 'strongly agree' to 'strongly disagree'. A typical item from this scale is "There is really no way I can solve some of the problems I have." This measure has been employed in studies of adults' coping with stressful life events. Another measure of psychological well-being was the one-item rating of happiness (Bradburn, 1969). This mea sure has been shown to correlate with other more lengthy measures of morale (Lawton, 1977}, and asks the subject "How would you say you are these days? Very happy, pretty happy, or not too happy?" The final psychological measure was the activities scale which listed 12 different activities ranging from card playing to going to the movies or meditating. Sub jects were asked to rate on a 5-point likert scale the degree to which they enjoyed the activity and to give an approximation of the number of times they had participated 37 in the activity during the previous month. One problem with past research in the field has been subjects' readi ness to change responses or to improve in order to please the experimenter (Felkins & Sime, 1981). Such a demand effect could threaten the internal validity of this study. Thus, the activity rating questions helped to fulfill sub jects' expectations of the questionnaire as a study of senior activities, and it was presumed to facilitate the subjects' impartiality in the study. Data from this mea sure were not included in the dependent variables eval uating the effect of the exercise program. 38 CHAPTER IV RESULTS The findings are presented in the following manner. First, the basic description and refinement of the dependent measures is discussed. Then mean values for the physiolog ical, cognitive and psychological variables are presented for each of the three groups at Time 1. This is followed by analyses of the more substantive research hypotheses. The analytic approach is summarized in a brief section, followed by a description of the pre-post results. These results are organized in subsections for the physiological, health, cog nitive, and psychological measures. A further subsection describes the relationships between areas of change. Next, analyses comparing demographic subgroups are presented, fol lowed by an overview of the Time 3 findings. Finally, the chapter concludes with a presentation of data describing the program drop-outs. Dependent Measures Correlations among variables in the three sets of mea sures are presented in Tables 4-6, and the means and stan dard deviations in Table 7, by group. On the cardiovascular measures, mean resting systolic 39 blood pressure ranged from 148 to 154, mean resting diastol ic blood pressure ranged from 84 to 93, and mean resting heart rate ranged from 70 to 79. While mean values for blood pressure were below the generally accepted cut-off for hypertension (160/95), approximately 40% of the sample was above the cut-off and more than 1/4 of the sample was taking medication for high blood pressure. Correlations among the cardiovascular measures, presented in Table 4, show that re lationships among the blood pressure measures are generally statistically significant, while among the heart rate mea sures there are also high intercorrelations. However, the correlations between heart rate measures and blood pressure measures are not consistently significant. Overall, these correlations would suggest that the resting and final dia stolic blood pressure measures are the best indicators of cardiovascular health. Correlations among the measures of sitting and stand ing flexibility, balance, and arm strength were moderate and no consistent pattern emerged from the relationships. More over, the variances of the balance and arm strength measures were extremely large (see Table 7). One explanation for this may lie in the difficulty of precisely standardizing these two measures. The same difficulty also pertained to the measure of standing flexibility. Consequently, it was determined that these three measures--balance, arm strength, and standing flexibility--would not be sensitive enough in- 40 Table 4 Intercorrelations Among Physiological Measures at Time 1 Resting Resting Resting HR SysBP Resting .11 HR Resting SysBP Resting DiaBP Stage HR Stage 2 HR Stage 3 HR Final SysBP Final DiaBP HR=Heart Rate SysBP=Systolic Blood Pressure DiaBP=Diastolic Blood Pressure * p<.OS ** p<.01 *** p(.001 DiaBP .27* .57*** Stage Stage 2 Stage 3 Final HR HR HR SysBP .72*** . 70*** .73*** .28 .17 .08 -.12 .67*** .36* .29 .13 .40* .80*** .80*** .42** .90*** .22 .13 Final DiaBP .43** .30* .83*** .49** .31 .38* .46** 41 dicators of physiological change, and they were dropped from further analysis. The sitting flexibility measure, because its administration entailed the use of a standardized ap paratus for measurement, appeared reliable and was retained in the set of dependent variables, along with weight and the cardiovascular measures. On the one item self-rating of health, 16% of the sam ple rated their health as excellent, 50% good, and 34% fair. No one in the sample rated health poor. This differs some what from results found for those over 65 years of age by the Department of Health, Education and Welfare who reported ratings of 29% excellent, 39% good, 22% fair, and 9% poor (U.S. Department of Health, Education, and Welfare, 1978). Thus, among the sample studied, the range of self-rated health was more restricted than among the HEW group~ over 80% rated their health as good or fair compared with 61% in the national survey. While there were no individuals who rated their health poor in this sample, there were also few who gave health an excellent rating. Mean scores on the Raven Progressive Matrices Test were between 20 and 22 which is approximately the median level for this age group (Raven, 1963). Mean levels for the Digit Span and Digit Symbol subtests were also within half a stan dard deviation of the median for this age group (Wechsler, 1981). Mean scores for the copying numbers test and the copying words test were both better than previously record- 42 ed scores for this age group (Birren & Botwinick, 1959). Correlations among the measures of fluid intelligence were consistently high, and all were statistically significant (see Table 5). The copy numbers test was reverse-scored for the intercorrelations, with higher scores indicating faster performance. Among the measures, Digit Symbol appeared to have the highest correlations with all of the other mea sures, suggesting that it may be a particularly strong in dicator of the construct. Mean scores on the Rotter I-E scale for other samples of community-residing older adults have been around 8 (Gatz, Siegler, George & Tyler, 1985}; thus, this sample appeared typical on this measure. Correlations among the psycholog ical well-being measures were moderate (Table 6}. Of par ticular interest, correlations among the three locus of con trol scales--Rotter locus of control, Lau and Ware health locus of control, and Pearlin mastery--were statistically significant but only about r=.3, indicating that they were indeed measuring different aspects of the construct. The health locus of control subscales did not appear to offer any distinctive information in relation to the other vari ables, and they were not analyzed separately. Happiness proved to be more allied with the symptom measures than with locus of control, but at a moderate enough level to suggest that the item makes independent con tribution to the domain of psychological well-being. 43 Table 5 Intercorrelations Among Fluid Intelligence Measures at Time 1 Raven' sa Digit Span Digit Symbola Copy Numbersb Copy Wordsa Raven's Digit Span .36** a Scored for n1Jilber oorrect Digit Symbol .62*** .58*** Copy Copy Numbers Words .40** .47*** .25* .50*** .49*** .80*** -.67*** b Scored for time taken to copy 1 page, then reversed such that higher soores indicate less time. * p{.05 ** p(.Ol *** p{.OOl 44 Table 6 Intercorrelations Among Psychological Measures at Time 1 CES- CES- Depression Anxiety CES- .73*** Depression CES Anxiety Rotter LOC Health LOC Pear lin Mastery Happiness Rotter LOC .21 .27 Health Pearl in LOC Mastery .45** .59*** .30* .38** .33* . 36** .26* Note: Scored for greater psychological distress, externality, and unhappiness * p<.OS ** p(.01 *** p(.001 Happiness .49** .41** .14 .30* .28* 45 For this sample, mean scores on the CES depression scale ranged from 10 among the social activity group to 18 among the exercise group. Epidemiological research sug gests that the most appropriate cut-off point for probable depressive disorder on the CES-D is 20 for older adults (Himmelfarb & Murrell, 1983). Thus, at Time 1 this sample was within the normal range though the exercise group was close to the cut-off value. ~Vhile 3/4 of the sample was below the cut-off value, the distribution across groups was not equal; approximately 70% of· the exercise and waiting list groups were below the cut-off, while all but one of the 15 individuals in the social activity group were below the cut-off level. Pretest Differences Among the Three Groups The mean values for the physiological, health, cogni tive, and psychological variables are provided in Table 7 for each of the three groups (exercise, C1, and C2) at Time 1. Analyses of variance were conducted to determine whether the groups differed initially in their mean level on any of these variables, i.e. whether randomization was successful. Significant main effects of group were found on three of the 36 variables; one health variable, satis faction with quality of sleep (F(2,40)=3.56, p<.OS), and two physiological variables, final systolic blood pressure (measured after the modified step-test) (F(2,29)=9.47, 46 Table 7 Pre-Test Mean Scores and Standard Deviations Ph!siolo~::ical Measures (li=48) Variable Exercise Group Social Activit! Group Waiting List Group (.H,.15) (H=15) (li=18) Mean S.D. Mean S.D. Mean S.D. Weight (kg.) 77.1 18.3 64.3 7.8 74.3 34.6 Flexibil(in.) 8.1 3.9 10.5 4.3 8.1 3.5 (Sitting) Flexibil (in.) 6.9 1.4 7.0 2.0 6.2 1.8 (Stand, left) Flexibil(in.) 6.7 1.0 7.1 1.1 6.1 1.4 (Stand, rt.) Balance(sec.) 5.1 4.3 11.0 12.4 11.5 20.7 (Lt. side) Balance(sec.) 5.6 4.2 9.5 8.4 12.0 18.4 (Rt. side) Strength(sec.) 72.3 47.6 66.1 39.5 86.5 45.3 (Rt. arm) Strength(sec.) 65.0 32.4 58.7 37.2 74.9 47.0 (Lt. arm) Resting HR 78.9 16.2 70.8 8.1 73.1 12.3 Resting SysBP 148.1 22.8 149.4 24.9 154.0 22.8 Resting DiaBP 93.1 14.8 84.5 11.7 92.4 12.0 Stage HR 89.7 11.8 78.8 7.6 84.4 12.5 Stage 2 HR 95.6 10.7 90.9 10.7 87.7 16.5 Stage 3 HR 100.6 14.7 99.6 12.9 94.3 19.7 Final SysBP 160.5 15.4 136.5 11.7 151.8 12.0 Final DiaBP 93.0 12.6 80.7 9.4 92.8 7.8 HRsHeart Rate SysBP•Systolic Blood Pressure DiaBP•Diastolic Blood Pressure 47 Table 7 (cont.) Pre-Test Mean Scores and Standard Deviations Cognitive Measures (N_,.,48) Variable Exercise Group Social Activity Group Waiting List Group (N.= 15) (~H5) (~= 18) Mean S.D. Mean S.D. Mean S.D. Copying(sec.) 99.3 19.5 128.8 88.8 118.3 44.6 Writing(// of 27.3 9.0 31.7 10.4 26.0 12.1 words) Digit Span 10.5 5.3 12.6 4.1 10.7 4.7 Digit Symbol 34.2 12.0 35.5 10.6 29.2 12.4 Raven's 21.3 10.3 20.7 6.3 20.2 7.6 Psvchological Measures CN.•48) Variable Exercise Group Social Activity Group Waiting List Group CN.~ 15) (H=15) CN.•18) Mean S.D. Mean S.D. Mean S.D. CES-Depres- 18.3 11. 1 10.7 9.3 14.3 11.6 sion CES-Anxiety 1.0 1.2 0.7 1.1 1.3 1.2 Health Locus 89.3 21.6 91.8 20.9 98.4 12 .o of Control Rotter Locus 9.2 3.9 5.9 4.4 8.8 3.0 of Control Pear lin 14.1 3.7 12.5 5.0 14.9 3.9 Mastery Happiness 1.6 0.7 1.9 0.5 2.0 0.7 Note: Higher scores indicate greater psychological distress, externality, and unhappiness. 48 Table 7 (cont.) Pre-Test Mean Scores and Standard Deviations Health Measures (li=48) Variable Exercise Group Social Activity Group Waiting List Group (R=15) Cli= 15) (R=18) Mean S.D. Mean S.D. Mean S.D. Alcohola 1.9 0.9 2.2 0.8 1.9 1.1 Smoking a 1.9 1.1 1.3 0.5 1.4 1.0 Diet a 3.6 0.5 3.0 1.0 3.1 0.7 Sleep a 3.2 0.9 2.5 1.1 3.4 0.8 Physical a 2.7 1.1 3.0 1.0 2.9 1.0 Exercise Activityb 1.2 1.2 1.1 1.4 1.0 1.4 (strenuous) Activityb 1.5 1.4 1.1 0.8 2.1 1.7 (moderate) Activitl 1.7 1.0 1.7 1.0 1.9 0.9 (overall) Healthc 2.0 0.8 2.3 0.7 2.2 0.7 a Scored with 1=low and 4=high bScored with O=low and S=high c1=excellent,2=good,3=fair,4=poor 49 p<.01}, and final diastolic blood pressure (F(2.29}=4.96, p<.OS}. Thus, the social activity group reported sleeping less well at Time 1 than the other two groups, and among this group blood pressure was less elevated after the step test than it was in the other two groups. While there was not a significant difference between groups on weight, the mean scores indicate that the exercise group was somewhat heavier than either of the other two groups. This has a bearing on measures of heart rate and blood pressure--es. pecially the elevated measures--since, in fact, the heavier individual must exert more force during the step-test to perform an amount of work equivalent to a lighter person. Thus, higher values would be expected for heavier individ uals on the measures of elevated heart rate and blood pres sure. There were no significant group differences for the remainder of the health and physiological variables nor for the psychological and cognitive variables at Time 1. Be cause two significant differences would have been expected by chance, it would seem that randomization was reasonably well achieved. It should be noted here and in further sec tions that the Hs and, consequently, the degrees of free dom vary due to missing data. In particular, those indi viduals whose resting blood pressure or heart rate were above the cut-off values did not have scores recorded for the three elevated heart rates or for final blood pressure. 50 Analysis of Pre-Post Change The remaining 30 dependent variables were analyzed using several analytic strategies. First, for the entire sample, a repeated measures analysis of variance (ANOVA) was conducted for each variable to determine main effects of Group and Time, as well as to determine any Group x Time interactions. Group was defined as exercise versus social activity versus waiting list. Time was defined as pre-test versus the first post-test. Pre-post repeated measures analyses of variance take into account the initial differ ences between individuals on the variables. A significant Group x Time interaction, with the exercise group improv ing relative to the two control groups, would serve to con firm the experimental hypo~~eses. Second, because there were relatively few males in the study, the repeated measures ANOVAs were again conducted using scores for only the females in the sample. An ANOVA of Time 1 scores for males and females indicated that fe males were significantly higher on the measure of depres sive symptomatology (F(1,38)=4.48, p<.OS) and were signifi cantly less flexible (F(1,42)=5.49, p<.OS) than the males. Thus, it was thought that unequal sex distribution within groups would attenuate any effects on these variables and that studying the females separately would constitute a better test of the hypotheses. Third, in order to increase the power of the tests, 51 and because drop-out was particularly severe in the social activities group, the control groups were pooled and the ANOVAs repeated. Thus, all of the analyses were conducted not only for the three groups--exercise, social activity, and waiting list--but also for the two groups formed when the control groups were pooled. For the latter pooled analyses the design of the study was altered. The social activities group was treated as a waiting list group and, thus, the experimental control for the social effects of the exercise classes was absent. Finally, correlated t-tests and chi-square tests were conducted to corroborate any trends evidenced by the ANOVAs. Mean values for all of the experimental variables for each of the three experimental groups at Times 1 and 2 are given in Table 8. This table includes values only for the 39 subjects tested at both Time 1 and Time 2. In Table 9 are provided mean values for all of the dependent variables for the females only in each of the three groups at Times 1 and 2. Results of the repeated measures analyses of var iance and all other analyses will be reported for four groups of variables: physiological, health, cognitive func tioning and psychological well-being. Physiological The three-group repeated measures analysis of variance 52 Table 8 Pre and Post-Test Mean Scores and Standard Deviations Physiological Measures (li=39) Variable Exercise Group Social Activity Group Waiting List Group @=14) (li=11) @=14) Time1 Time2 Time1 Time2 Time1 Time2 Mean Mean Mean Mean Mean Mean (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) Weight (kg.) 77.1 76.0 63.6 63.0 69.8 70.1 (19.0) ( 18. 8) (8.5) (8.3) (19.0) (19.3) Flexibil (in.) 8.1 9.1 10.5 9.1 9.2 8.7 (Sitting) (4.0) (4 .1) (4 .8) (5. 5) (2. 7) (2.2) Resting HR 79.2 79.9 70.5 68.3 74.0 72.7 (16. 8) (12.6) (7.6) (9.9) (13 .0) (10. 2) Resting SysBP 148.7 151.3 151.4 142.0 154.6 144.3 (23. 6) ( 15. 7) (22.5) (22.8) (20.0) (16.9) Resting DiaBP 92.7 87.6 85.6 79.0 93.0 86.9 (15. 3) (15. 8) (13.5) (11.0) (12. 9) (11. 7) Stage HR 89.1 98.6 77.3 78.8 87.3 84.3 (12. 2) (16.4) (8.6) (11.5) (12.6) (13.8) Stage 2HR 95.3 98.7 90.3 83.0 91.0 90.0 (11.4) (10 .5) (12.1) (12.3) (19. 7) (17.6) Stage 3HR 101.5 103.4 96.2 90.3 97.5 98.1 (15. 8) (6.3) (12. 8) (16. 9) (23.8) (16.5) Final SysBP 162.0 160.4 136.9 145.9 153.1 147.7 (15.3) (27.3) (11.8) (21.6) (13.2) (22. 7) Final DiaBP 92.2 88.8 80.0 83.6 93.1 85.8 (12.9) (8. 6) (10. 6) (10. 5) (8. 6) (12.2) HR=Heart Rate SysBP•Systolic Blood Pressure DiaBP•Diastolic Blood Pressure 53 Table 8 (cont.) Pre and Post-Test Mean Scores and Standard Deviations Cognitive Measures Q!=39) Variable Exercise Group Social Activity Group Waiting List Group (!i= 14) ([=11) (!i= 14) Time! Time2 Time! Time2 Time! Time2 Mean Mean Mean Mean Mean Mean (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) Copying(sec.) 100.8 113.6 143.0 109.0 111.7 116.0 (19.3) (8.6) (100.4) (33.8) (29. 5) (28.8) Writing(/! of 26.8 26.0 30.4 29.7 25.6 26.0 words) (9. 1) (10.2) (10.6) (11.3) (10.9) (9.9) Digit Span 10.6 11.5 12.3 12. 1 10.6 10.8 (5.5) (4 .3) (3.4) (3.8) (4.4) (4 .5) Digit Symbol 34.4 35.8 35.4 36.2 28.5 29.8 (12. 5) (12. 6) (9. 7) (10. 2) (12.8) (11.8) Raven's 21.1 21.0 21.0 22.8 20.4 20.7 (10.7) (8.0) (6. 1) (8. 9) (7.6) (8.3) Psychological Measures (li=39) Variable Exercise Group Social Activity ~ Waiting List Group (.li•14) (Ji•11) (li:o14) Time1 Time2 Time1 Time2 Time1 Time2 Mean Mean Mean Mean Mean Mean (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) CES-Depres- 18.9 17.3 10.8 10.7 12.5 15.7 sion (11.5) (9.0) (9.8) (7.8) (10. 5) (8.6) CES-Anxiety 1.1 0.8 0.8 0.9 1.2 1.2 (1.3) (0 .8) ( 1.1) ( 1.1) ( 1.1) (0.9) Health Locus 88.9 91.7 89.7 100.2 98.3 96.8 of Control (22.4) (17.3) (22.0) (22. 4) (13 .1) (11.6) Rotter Locus 8.8 8.7 5.7 6.2 8.9 9.8 of Control (3. 7) (3.4) (4. 0) (3.4) (2. 6) (3 .1) Pear lin 14.5 14.8 11.8 13.3 13.8 15.4 Mastery (3.5) (4.0) (4. 7) (5. 3) (3 .2) (3. 8) Happiness 1.5 1.7 1.9 2.0 1.9 2.1 (0. 7) (0. 7) (0. 6) (0.8) (0.6) (0. 8) Note: Higher scores indicate greater psychological distress, externality, and unhappiness. 54 Table 8 (cont.) Pre and Post-Test Mean Scores and Standard Deviations Health Measures ---(li,=39) Variable Exercise Group Social Activity Group Waiting List Group (,li= 14) (!!.=11) (N.= 14) Time1 Time2 Time1 Time2 Time1 Time2 Mean Mean Mean Mean Mean Mean (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) Alcohol a 2.0 1.9 2.3 2. 1 2.1 1.9 (0. 9) (0. 8) (0 .8) (0. 9) ( 1.1) (0. 9) Smoking a 1.9 2.3 1.3 1.2 1.5 1.6 ( 1. 1) (1.2) (0.5) (0.4) (1.1) (1. 2) Diet a 3.6 3.2 3.1 3.5 3.1 3.3 (0.5) (0. 9) ( 1.1) (0.5) (0. 7) (0. 7) Sleep a 3.2 3.3 2.4 2.8 3.4 3.2 (1.0) (0. 9) ( 1. 2) ( 1.1) (0. 9) (0. 6) Physical 2.9 3.5 3.2 3.5 3.0 3.0 Exercise a ( 1. 0) (0. 8) (0 .8) (0. 7) (0. 9) (0. 9) Activity b 1.3 2.4 1.0 1.2 1.1 1.1 (strenuous) ( 1.2) ( 1. 9) (1.4) ( 1. 3) ( 1. 4) ( l. 5) Activity b 1.6 1.7 1.2 1.8 2.4 1.2 (moderate) (1.4) (1. 3) (0. 8) (1.5) ( 1.6) (1.4) Activity a 1.8 2.0 1.8 1.8 2.1 1.7 (overall) ( 1.0) (0. 9) (1.1) (0.8) (0. 9) (0. 7) Healthc 2.0 2.4 2.3 2.5 2.3 2.4 (0. 8) (0 .6) (0. 7) (0. 8) (0. 6) (0.5) aScored with 1=low and 4=high bScored with O=low and 5•high c1•excellent,2=good,3=fair,4=poor 55 Table 9 Pre and Post-Test Mean Scores and Standard Deviations for Females Physiological Measures (li~34) Variable Exercise Croup Social Activity Group Waiting List Group (N=13) @=8) (ll,=13) Time1 Time2 Time1 Time2 Time! Time2 Mean Mean Mean Mean Mean He an (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) Weight (kg. ) 77.2 76.0 61.4 61.8 70.5 70.8 (19.8) (19.5) (9.0) (9. 3) (19.6) (19.9) Flexibil(in.) 8.6 9.3 12.3 11.5 9.1 8.7 (Sitting) (3. 8) (4. 1) (3.9) (4. 1) (2.8) (2. 3) Resting HR 80.4 80.8 70.4 66.8 72.3 72.9 (16.8) (12. 7) (4. 9) (8. 7) (11.8) (10. 6) Resting SysBP 150.2 152.2 149.3 137.1 156.5 146.0 (23.9) (16. 0) (24.8) (25.4) (19.5) (16.3) Resting DiaBP 93.5 88.2 84.0 76.0 94.0 88.6 (15. 6) (16.3) (14. 9) (8.8) (12. 9) (10. 0) Stage 1 HR 90.1 100.7 79.8 78.0 86.0 85.7 (12. 4) (16.5) (5 .8) (10. 2) (13.3) (13.5) Stage 2 HR 97.1 100.5 94.0 84.0 88.7 92.3 (10. 7) (10. 2) (7 .8) (9.8) (23.4) (17 .4) Stage 3HR 104.6 105.8 100.6 91.8 92.7 97.3 (15. 5) (4 .1) (7.6) (16. 3) (92.7) (17 .4) Final SysBP 164.4 160.4 136.0 137.7 155.3 150.8 (13.7) (29.5) (12. 4) (17.1) (13.0) (20.5) Final DiaBP 93.6 90.0 79.4 78.7 92.7 87.1 ( 12. 7) (8.5) (11.3) (5. 9) (9. 4) (11. 7) HR•Heart Rate SysBP•Systolic Blood Pressure DiaBP•Diastolic Blood Pressure 56 Table 9 (cont.) Pre and Post-Test Mean Scores and Standard Deviations for Females Variable Exercise Group (li=13) Time1 Time2 Mean Mean (S.D.) (S.D.) Copying(sec.) 100.3 (20.0) 112.5 (33.2) Writing(# of 27.1 words) (9.4) Digit Span 10.9 (5. 6) Digit Symbol 34.5 (13.0) Raven's 20.5 (10. 9) 26.5 (lO.l•) 11.6 (4 .5) 36.1 (13.1) 21.5 (8. 1) Variable Exercise Group (lis 13) CES-Depres sion CES-Anxiety Health Locus of Control Rotter Locus of Control Pear lin Mastery Happiness Timel Time2 Mean Mean (S.D.) (S.D.) 19.5 (11.8) 1.2 (1. 3) 90.2 (22. 7) 9.4 (3.3) 14.8 (3.4) 1.5 (0. 7) 17.7 (9 .2) 0.8 (0 .8) 94.2 ( 15. 1) 9.0 (3.4) 15.3 (3.8) 1.8 (0. 7) Cognitive Measures (li=34) ~ Activity Group (liaS) Time1 Time2 Mean Mean (S.D.) (S.D.) 122.5 (70.3) 32.8 (9. 3) 12.6 (3.8) 37.8 (9. 5) 22. 1 (5.5) 91.6 (15. 8) 32.4 (10. 4) 12.3 (4.3) 39.4 (6. 8) 25.5 (6. 6) Psychological ~easures (N=34) ~ Activity Group Cli=B) Time1 Time2 Mean Mean (S.D.) (S.D.) 8.3 (6. 8) 0.4 (0. 8) 86.2 (22.6) 5.5 (4. 3) 10.9 (3.9) 1.9 (0. 4) 10.3 (6.0) 0.6 (0. 9) 98.3 (24.8) 5.8 (2.6) 12.3 (4. 7) 1.9 (0.6) Time1 Mean (S.D.) 110.9 (30. 7) 25.6 (11.4) 10.8 (4.5) 29.4 (12.9) 20.8 (7.8) Time2 Mean (S.D.) 116.2 (30. 3) 26.2 (10.5) 11.0 (4. 6) 30.3 (12. 2) 21.3 (8.4) Waiting List Group Qi=13) Time1 Mean (S.D.) 10.9 (9. 1) 1.1 ( 1.2) 97.8 (13.5) 8.9 (2.6) 13.8 (3.2) 1.8 (0. 6) Time2 Mean (S.D.) 14.9 (8.5) 1.1 (0. 9) 97.6 (11. 7) 9.5 ( 3. 1) 15.0 (3.5) 2.0 (0. 7) ~ote: Higher scores indicate greater psychological distress, externality, and unhappiness. 57 Table 9 (cont.) Pre and Post-Test Mean Scores and Standard Deviations for Females Health Measures -cli=34) Variable Exercise Group Social Activity Group Waiting List Group CN-13) (.!i=8) (li=13) Time1 Time2 Time1 Time2 Time1 Time2 Mean Mean Mean Mean Mean Mean (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) Alcohol a 2.0 1.8 2.4 2.1 2.0 1.9 (0.9) (0. 7) (0.8) ( 1.0) (1.2) (1.0) Smoking a 2.0 2.3 1.3 1.1 1.5 1.6 ( l. 1) ( l. 3) (0.5) (0. 4) (1.2) ( 1.2) Diet a 3.6 3.2 3.0 3.5 3.2 3.3 (0.5) (0. 9) ( l. 2) (0 .5) (0. 7) (0.8) Sleep a 3.2 3.3 2.3 2.9 3.3 3.2 (1.0) (1.0) (1. 1) ( l. 1) (0. 9) (0 .6) Physical 2.8 3.5 3.3 3.6 3.2 2.9 Exercise a ( 1.1) (0. 8) (0 .8) (0 .5) (0. 7) (0. 9) Activity b 1.3 2.2 1.3 1.5 1.2 1.2 (strenuous) ( l. 2) (2.0) (1.5) ( l. 4) (1.5) (1.5) Activity b 1.6 1.6 1.1 1.8 2.6 1.3 (moderate) (1.4) (1.3) (0. 8) ( 1.8) (1.5) (1.4) Activity a 1.7 1.9 1.7 1.9 2.2 1.8 (overall) (1.0) (0. 9) ( l. 1) (0.9) (0. 8) (0. 7) Healthc 2. 1 2.5 2.1 2.4 2.2 2.5 (0. 8) (0.5) (0. 7) (0. 7) (0 .6) (0 .5) aScored with 1=low and 4•high bScored with O=low and 5=high c1•excellent,2=good,3=fair,4=poor 58 (ANOVA} revealed no significant main effects for Time or for Group on weight, flexibility, the blood pressure mea surements, or the heart rate measurements. There were also no significant Time x Group interactions. When the control groups were pooled, repeated measures ANOVA indicated a main effect of Group on resting heart rate (F(1,37)=4.54, p<.05), on stage 1 heart rate (F(1,18) =5.25, p<.05), and on stage 2 heart rate (F(1,12)=7.03, p<.05). Thus, the heart rate measurements for the exer cise group were consistently higher than for the combined control groups at both times of measurement, suggesting that the exercise group may have been less physically fit than the control groups both at the beginning of the pro gram and at the end of the first 12 weeks. There was also a significant Time x Group interaction on the measure of resting systolic blood pressure (F(1,37)=5.87, p<.OS) with the exercise group remaining at the same level between Times 1 and 2, while systolic blood pressure dropped among the controls. Other analyses with the control groups pooled suggest ed that exercise benefitted flexibility and weight. There was a trend toward a Time x Group interaction on the mea sure of sitting flexibility (~(1,35)=3.10, p<.09), which was substantiated in a subsequent chi-square test of flex ibility where the exercise group was found to be signifi cantly more likely to have increased one inch or more (chi- 59 square(2)=8.44, p<.OS). When data for the females only were analyzed, ANOVA revealed no significant main effects, and there were no significant Time x Group interactions. However, for fe males only, when the control groups were pooled, the ANOVA indicated a significant Time x Group interaction on weight (F(1,31)=9.66, p<.OS), reflecting a weight loss among the exercise group while the control groups maintained the same weight between Time 1 and Time 2. Overall, the results for the physiological measures were mixed. While the cardiovascular measures used in this study did not change significantly among the exercise group during the first 12 weeks, the significant improvements in measures of flexibility and weight for some members of the exercise group indicated a physiological effect that was not observed among either of the two control groups. How ever, the relatively small magnitude of these changes would suggest that factors other than the exercise effect (e.g. motivational factors or random fluctuations in weight and flexibility) may explain the observed differences. This is particularly relevant for the measure of weight since weight loss may have resulted from reduced water retention, from reductions in fat or, in fact, the people who were motivated to exercise may also have been motivated to im prove their diets. Furthermore, the significant effect found for the measure of flexibility represented not only 60 a gain for the exercise group, but also a loss of flexibil ity for the two control groups. Thus, there was only par tial support for the experimental hypotheses concerning weight and flexibility, while hypotheses concerning heart rate and blood pressure were not supported. Health The three-group repeated measures ANOVA for the health variables revealed no main effects. However, for the ac tivity variables there was a significant Time x Group in teraction on the measure of level of involvement in moder ate physical activity (suchasdancing, gardening, working with home tools): (F(2,32)=3.58, p<.OS). The social activ ities group showed the greatest increase, while the wait ing list control group evidenced a decrease. Among the females only, repeated measures ANOVA indi cated a significant Time x Group interaction on the measure of perceived participation in physical exercise (F(2,29)= 3.35, p<.OS); in contrast to the control groups, the exer cise group reported an increased level of exercise. No other significant main effects or interaction effects on any health variables were found among the females. Follow-up correlated t-tests with both males and fe males also indicated increased self-reported participation in physical exercise on the part of the exercise group be tween pre and post-test (i(13)=3.23, p<.01). This finding 61 was further supported by a trend on the chi-square test comparing those whose self-ratings increased by 1 point or more on the 4 point likert scale to those who indicated no change {chi-square{2)=5.05, p<.09). Thus, subjects in the exercise condition were likely to perceive themselves as more active in physical exercise at Time 2 than at Time 1, while moderately strenuous activ ity increased for the social activities group. This find ing may be, to a large extent, a reflection of subjects' participation in the exercise class or non-physical activ ity, respectively. However, these findings serve to verify that participants' perceptions of level of physical activ ity changed in the hypothesized direction. No other health behaviors--including diet, sleep, alcohol consumption, and smoking behavior--appeared to change overall or differ entially as a result of the intervention. Cognitive Functioning No significant main effects were found on any of the cognitive measures with the three-group repeated measures ANOVA. A Time x Group trend was found on the copying num bers measure {F(2,34}=3.03, p<.07}, with the exercise group and the waiting list group performing more slowly at Time 2 than at Time 1, while the social activity group improved. This interaction reached significance for the females only (F(2,29}=3.71, p<.OS}. One possible explanation for the 62 scores of the social activity group lies in the performance of one individual who took an extremely long time at Time 1 and substantially reduced her score at Time 2. When the repeated measures ANOVA was conducted without this outlying score, the interaction was eliminated. No other signifi cant interactions were found for the cognitive measures, and subsequent chi-square tests and paired t-tests con firmed no between groups effect of the intervention on these measures of cognitive functioning. Despite the absence of the hypothesized group effects for measures of fluid intelligence, attributable in part to inadequate statistical power, there were some suggestive relationships between physiological and cognitive change, reported below in the section following psychological well being. Psychological Well-Being The three-group repeated measures ANOVA comparing the groups at pre and post-test revealed no significant main effects of interactions for any of the measures of psycho logical well-being. When control groups were pooled, ANOVA indicated a trend toward a Time x Group interaction (F(1,32)=2.88, p<.10) on depression, which reached the established statistical significance level in the analysis of women only with control groups pooled (F(1,28)=4.22, p<.OS). Thus, participants in the exercise group, espe- 63 cially females, indicated fewer symptoms of depression at post-test than at pre-test, while levels of depressive symptoms either remained relatively unchanged or increased among the control groups. At both pre and post-test, the exercise group was more depressed than the other two groups. While the ANOVA corn paring Time 1 scores did not register a significant dif ference, for females only the pattern was particularly pro nounced. In the females only ANOVA, there was a signifi cant main effect of group on depression. In other words, at both Time 1 and Time 2 the women in the exercise group indicated higher levels of depressive symptoms than the women in the other two groups. Pre-test mean scores ranged from 8 in the social activity group to 19 in the exercise group. Because the exercise group began the program with the highest level of depressive symptomatology, reduced symptoms among this group may represent regression toward the mean as well as a response to the exercise program. For both males and females, although mean scores for the exercise group decreased between pre and post-test, the proportion of subjects who were clinically depressed (score > 19} remained at approximately 35% between tests. Similarly among the social control group there was no in crease of depressed individuals; but among the waiting list group the proportion jumped from 23% to 46%. In addition to this somewhat mixed picture of reduced 64 depression among the exercise group, there was some evi dence of reduced anxiety. Because the ANOVA was not sensi tive enough for use with one item, a chi-square test was used to compare those who decreased 1 point or more on the measure of self-rated anxiety to those who did not change. With the control groups pooled, the analysis indicated a trend on this measure (chi-square(1)=3.59, p<.06), with mean level of anxiety decreasing nearly 1/2 point on the 4 point scale among the exercise group between Times 1 and 2, while remaining stable among the controls. Thus, the self-rated symptoms of psychological dis tress--both depression and anxiety--appeared to be respon sive to the exercise intervention, and there was partial statistical support for the experimental hypotheses. Hypo theses concerning the measures of locus of control, mas tery, and happiness were not confirmed, although the mean scores on both the Rotter I-E Scale and the Pearlin mastery items changed in the expected direction. Again, these findings are further elucidated by examining relationships between physiological and psychological change. Relationship Between Areas of Change In order to examine more fully the relationships be tween each of the three clusters of variables, several analyses were conducted. In particular, these helped de scribe further the relationship between the physiological 65 measures and the cognitive or psychological measures in the study. First, change scores were computed by subtracting each individual's pre-test score from his or her post-test score (with direction of scoring changed to make positive scores indicate improvement); then, correlations between change scores were computed for participants in the exer cise group. There were a total of 50 correlations, ranging from r=.83 to r=-.45 (mean r=.17). Seven were statistical ly significant, more than would be expected by chance. These correlations indicated that improvement between pre and post-test on the writing numbers test was significant ly correlated with weight. loss (r=.53, p<.05) and with de creased resting diastolic blood pressure (E=.65, p<.01). Copying words was also significantly correlated with these physiological measures (r=.55, p<.05 and £=.48, p<.05, re spectively). Furthermore, improved scores on the Digit Span and Digit Symbol subtests of the WAIS were correlated with lowered heart rate on Stages 1 (r=.76, p<.05) and 2(r=.83, p<.05), respectively, of the modified step-test; and improvement on the Raven's was associated with de creased final systolic blood pressure (r=.73, p<.05). Thus, these results provide evidence that, in fact, among the exercise group improvement on physiological indicators was associated with improvement on measures of fluid intel ligence. Similar correlations were calculated between physic- 66 logical change and improved psychological well-being. There were a total of 60 correlations (mean r=.20). Physi ological improvement was significantly associated with in creased sense of mastery and sense of internal locus of control. Decreased resting diastolic blood pressure was correlated with increased levels of mastery (r=.65, p<.01), and decreased final systolic blood pressure was associated with greater internal locus of control (r=.77, p<.03). Furthermore, decreased final systolic blood pressure was also associated with lowered levels of anxiety (r=.69, p<.05). Thus, gains in physiological status were associat ed with increases in perceived control or mastery and with decreases in anxiety. Finally, those who attended 75% or more of the classes were compared to less regular attenders, using chi-square, to test whether greater exposure to the intervention led to more improvement on measures of psychological well-being. Indeed, greater attendance at exercise classes was associ ated with increased levels of mastery, i.e. a gain of 1 point or more on the mastery scale (chi-square(1)=4.72, p<. 05) • Thus, while many of the hypothesized differences be tween the three groups in this experiment did not reach statistical significance, it is useful to note that among the exercise group, physiological improvement was associ ated with improvement on measures of fluid intelligence as 67 well as on measures of mastery, locus of control, and anx iety. Furthermore, greater attendance at the exercise classes was more strongly associated with increases in per ceived mastery. Results for the measure of depressive symptomatology warrant special attention. While improvement in cardio vascular status was associated with greater perceived con trol or mastery and lowered anxiety, there were no signi ficant correlations between cardiovascular improvement and decreased depressive symptoms. Furthermore, there was a negative correlation between the flexibility measure and depressive symptoms (r=-.61, p<.OS) suggesting that the exercise participants who improved on the measure of flex ibility were not the same individuals who indicated lowered depressive symptoms at post-test. While this finding is surprising, there are a couple of possible explanations for it. First, the mean depression score at Time 1 for the group that improved on flexibility was significantly lower (X=12.5) than the mean score for the group that did not change on flexibility (X=28.4; t(9)=2.78, p<.OS). Thus, decreases in levels of depressive symptoms among the group that began the program with greater depressive symptoms may have been due to regression toward the mean. Alternative ly, another explanation for the observed pattern of cor relations may be that the mechanisms underlying change on depressive symptoms are different from the mechanisms un- 68 derlying changes in locus of control or mastery. Analysis of Demographic SUbgroups Given the relatively older age range of these subjects and the ethnic diversity of the sample, it was pertinent to examine further the relationships between the dependent variables and age or ethnic status. In order to look at age-related effects on the change from pre to post-test, correlations between age and difference scores for the de pendent variables were calculated. Twenty-six correlations were calculated including the physiological, cognitive, and psychological well-being measures, as well as the measure of physical exercise. Correlations ranged from .26 to -.44 with a mean of -.03. Only three of the correlations reached the level of statistical significance; two cardio vascular measures, final systolic blood pressure (r=-.42, E<.OS), and final diastolic blood pressure (E=-.42, p<.OS), as well as one measure of fluid intelligence, Raven's (r=-.44, p<.01). These results indicate that younger age was associated with greater improvement on the measure of final blood pressure and on the Raven's test. Because the N for each distinct ethnic or racial group would be too small for meaningful statistical analysis, separate repeated measures ANOVAs were used to study the effects among Anglos only and among minorities only, with minorities including Blacks, Hispanics, and Asians. The 69 exercise intervention appeared to benefit different vari ables for Anglos and minorities. For minorities, repeated measures ANOVA, with the control groups pooled, indicated that there was a significant Time x Group interaction on the measure of depressive symptomatology (F{1,17)=4.56, p<.OS). Thus, mood improved for minority exercise parti cipants. On the other hand, among Anglos the exercise in tervention appeared to affect mainly physiological indi cators. There was a significant interaction on the measure of sitting flexibility {F{1,14)=6.82, p<.OS) with the exer cise group becoming more flexible while the control groups remained relatively stable; and there was a trend toward an interaction on weight {F(1,14)=3.96, p<.07) with the exer cise group losing weight while the controls remained the same. Time 3 Follow-Up A total of 29 subjects chose to continue in the exer cise program after Time 2, for another 12 weeks; 11 of the 14 subjects from the exercise condition, 8 of the 11 sub jects from the social activity condition, and 10 of the 14 subjects from the waiting list. Mean scores for all of the experimental variables for each of the three experimental groups at Times 1, 2 and 3 are given in Table 10. For the exercise group, it was of interest to examine whether continuing in an exercise program for 24 weeks led 70 Table 10 Mean Scores and Standard Deviations for Time 3 Follow-up Phzsiolo8ical Measures (li~29) Variable Exercise Group Combined Control Groups (li= 11) Cl:!.=18) Time1 Time2 Time3 Time2 Time3 Mean Mean Mean Mean Mean (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) Weight (kg.) 78.0 77 .o 76.8 66.8 67.2 (17.8) (16.3) ( 15. 7) (17.8) (17.5) Flexibil (in.) 8.1 9.8 9.6 7.9 8.6 (Sitting) (3.6) (3. 4) (3.9) (3.4) (3.4) Resting HR 81.0 80.5 78.9 70.8 73.3 (16. 0) (12. 0) (11.2) (9. 8) (9.5) Resting SysBP 151.5 154.0 147.3 145.9 142.0 (25.3) (14. 7) (20. 7) (19.0) (20. 6) Resting DiaBP 94.4 88.5 86.5 84.1 83.2 (15. 3) (10.9) (9. 1) (12. 7) (12. 4) Stage 1 HR 88.0 93.0 90.1 83.8 91.1 (13.0) (8.4) (9. 0) (13.6) (19.5) Stage 2HR 95.0 100.0 93.8 86.8 87.8 (12 .5) (11.8) (3.4) (16. 9) (19.3) Stage 3 HR 97.8 103.3 102.0 96.6 97.3 (14. 5) (8.6) (7.2) ( 14. 3) (20. 6) Final SysBP 162.0 153.7 152.7 147.4 147.0 (17 .0) (14.3) (21. 9) (22.4) (19.6) Final DiaBP 90.9 86.7 80.3 86.6 84.4 (14. 0) (8.5) ( 10. 2) (12.9) (8. 8) HR=Heart Rate SysBP=Systolic Blood Pressure DiaBPKDiastolic Blood Pressure 71 Table 10 (cont.) Mean Scores and Standard Deviations for Time 3 Follow-up Variable Exercise Group Q!.=- 11) Time1 Time2 Mean Mean (S.D.) (S.D.) Copying(sec.) 96.8 106.7 ( 16. 3) (28. 3) Writing (II of 28. 1 27.5 words) (9. 1) (10. 1) Digit Span 11.4 12.4 (6. 1) (4.5) Digit Symbol 36.5 37.9 (9.9) (11.6) Raven's 23.2 22.2 (9. 9) (7.9) Variable Exercise Group Q!.= 11) Time1 Timc2 Mean Mean (S.D.) (S.D.) CES-Depres- 14.9 15.0 sion (6. 1) (8,4) CES-Anxiety 0.8 0.9 ( 1. 0) (0. 8) Heal::h Locus 80.8 87.2 of Control (17 .4) (16.8) Rotter Locus 8.5 8.6 of Control (!1.1) (3,7) Pear lin 14.7 14.4 Mastery (3.4) (4.2) Happiness 1.4 1.6 (0. 5) (0, 7) Cognitive Measures <1:!.=29) Time3 Mean (S.D.) 101.5 (23.2) 28.5 (9.6) 10.9 (6.0) 37.8 (13 .1) 23.3 (6 .1) Combined Control Groups <1:!.=18) Time2 Time3 Mean Mean (S.D.) (S.D.) 113.4 138.7 (33.6) (89. 6) 27.9 27.5 (10. 9) (10 .5) 11.1 9.6 (4 .0) (3.9) 31.9 33.3 (10. 4) (10. 8) 20.1 21.6 (8.3) (6.0) Psychological Measures <1:!.•29) Combined Control Groups ([•18) Time3 Time2 Time3 Mean Mean Mean (S.D.) (S.D.) (S.D.) 15.7 13.3 12.7 (8.4) (9.5) (8.5) 0.8 1.1 1.2 (1.0) ( 1.1) (1.0) 85.9 99.7 99.0 (15. 9) (16,6) (17.4) 8.9 8. 1 8. 1 (3.8) (3.4) (3. 6) 14.2 14. 1 14.9 (4 .2) (4. 7) (4. 2) 1.8 1.9 1.8 (0 .8) (0. 8) (0. 9) 1<ote: Higher scores indicate greater psychological distress, externality, and unhappiness. 72 Table 10 (cont.) Mean Scores and Standard Deviations for Time 3 Follow-up Health Measures ---ai=29) Variable Exercise Group Combined Control Groups (]:-11) <.~-18) Timel Time2 Time3 Time2 Time3 Mean He an Mean Mean Mean (S.D.) (S.D.) (S.D.) (S.D.) (S.D.) Alcohol a 1.8 1.9 2.0 2.2 2.4 (0. 9) (0. 8) (0.8) (0. 9) (1.0) Smoking a 2.0 2.2 2.1 1.4 1.4 (1.2) (1.2) ( 1. 4) (1.0) (1.0) Diet a 3.6 3.4 3.2 3.3 3.3 (0.5) (0. 7) (0. 8) (0. 7) (0. 7) Sleep a 3.0 3. 1 2.8 2.9 3.1 (1.0) (1.0) ( 1.0) (0. 9) (0. 8) Physical 2.9 3.5 3.4 3.1 3.3 Exercise a ( 1. 1) (0 .8) (0. 7) (0. 9) (0 .8) A . . b ct1v1ty 1.3 2.3 2.6 1.0 1.4 (strenuous) (1.2) (2. 1) ( 1. 6) ( 1. 2) (1. 7) Activity b 1.6 1.9 1.8 1.4 2.2 (moderate) ( 1. 6) (1. 3) (1.5) (1.4) ( 1. 8) Activitya 1.9 2.1 2.1 1.8 2.1 (overall) (1. 0) (0.8) (1. 0) (0. 8) (0 .8) Healthc 2.1 2.4 2.4 2.5 2.3 (0.8) (0. 7) (0.8) (0. 6) (0.5) aScored with 1=low and 4=high bScored with O .. low and Sahigh cl .. excellent,2•good,3 .. fair,4"'Poor 73 to improvements not apparent after only 12 weeks. Time series analysis of the dependent measures among the exer cise group across the three times of measurement indicated a significant effect for the measure of sitting flexibility (F(2,20)=4.65, p<.OS), due mainly to the change from Time 1 to Time 2. There was also a trend toward a significant effect on the measure of final diastolic blood pressure (F(2,6}=5.64, p<.07}. Both final and resting diastolic blood pressure, as well as resting heart rate, continued to improve Time 2 to Time 3, although the latter two mea sures not significantly. Thus, the exercise group appeared to improve physiologically. However, no trends among the cognitive, health or psychological well-being measures reached statistical significance. One possible explanation for minimal continued improvement between Time 2 and Time 3 could be the fact that attendance was less steady during this follow-up period. While close to 90% of the exercise group attended more than 3/5 of the classes during the first 12 weeks, only approximately 60% achieved 3/5 at tendance during the second twelve weeks. Paired t-tests were conducted for scores of the pooled control groups comparing Time 2 to 3. This analysis con stituted a replication of the effects of the exercise group Time 1 to Time 2. These results indicated no significant changes in mean scores on any of the dependent variables. However, attendance among this group was also a problem~ 74 during the twelve week period less than 60% attended 3/5 of the classes. Adherence and Analysis of Drop-Outs The literature on exercise adherence suggests that drop-out rates observed in clinical exercise settings ap proach or exceed 50% within the first 3-6 months of initial involvement (Dishman, 1982). In this study, only 7% dropped out of the exercise class during the first 12 weeks and attendance for those who remained ranged from 61% to 94% of classes. In fact, 90% of the exercise group attend ed at least 3/5 of the sessions, and 60% attended at least 3/4 of the sessions. These results for adherence and at tendance compare quite favorably to those reported else where in the literature (Gale, Eckhoff, Mogel & Rodnick, 1984). During the second 12 weeks of exercise an additional 20% dropped out of the exercise group, while 28% of the combined control groups dropped--making a total of 26% who dropped out during the second phase of the program. Over all, 40% of the group tested at Time 1, including exercise as well as the two controls, had dropped from the program by Time 3. An interesting aspect of this study, though it was not one of the original experimental hypotheses, was the dif ference between drop-outs from the program and those who 75 attended on a number of the experimental variables. In or der to explore this phenomenon, t-tests for independent groups were computed for the mean scores at Time 1 of those who attended and were tested at Time 2 versus those who dropped out sometime before Time 2. These were computed for the whole group as well as within each of the three conditions. For the total group, those who attended through Time 2 reported that they took part in significantly more physical exercise at Time 1 than those who dropped out (t(39)=2.66, p<.05). Thus, it was the least active who chose not to participate in the program. Within the exercise group, the one individual who dropped out was less happy than the rest of the group and her self-rated happiness was more than two standard devia tions below the group mean. Her rating of physical exer cise was more than one standard deviation below the mean. "Drop-outs" from the waiting list group indicated sig nificantly lower levels of mastery (t(13)=2.96, p<.05) and were significantly less flexible at Time 1 (t(16)=2.77, p<.05) compared to waiting list subjects who stayed in the project. Of the four individuals who dropped from the waiting list, three were male. These males indicated gen erally greater psychological distress than attenders; their scores on the measures of happiness, anxiety, depression, and mastery were more than one standard deviation below 76 the mean. Thus, while males with higher than average de pressive symptoms dropped out of the program early, females with similarly high depressive symptoms took part in the program for at least the first 12 weeks. Among the social activity group, there were no dif ferences on the dependent variables between those who at tended and those who dropped out. Further t-tests were then conducted comparing scores of those who attended through Time 3 versus those who dropped out sometime before Time 3 (including those who dropped out before Time 2). Among the exercise group it was found that drop-outs were significantly more external on health locus of control (t(13)=3.27, p<.01). Further more, drop-outs indicated significantly greater symptoms of depression (t(11)=2.39, p<.OS) and they were less happy (t(13)=2.38, p<.OS). Thus, the exercise drop-out was char- acterized by lower self-rated happiness (all drop-outs were at least one standard deviation below the mean) and greater externality on health-related beliefs (scores for the three who dropped out were more than one standard deviation above the mean). While the t-test indicated that depressive symptoms were significantly greater among the drop-outs, it appears that this was due to the extremely elevated scores of two of the three women who dropped out during the second twelve weeks. There were no consistent patterns among the various physiological measures or health behaviors for ex- 77 ercise drop-outs. Among the social activity group drop-outs reported having a less well-balanced diet than did those who attend ed (t(10)=2.90, p<.OS). In contrast to the patterns with other groups, the three drop-outs from this group between Time 2 and Time 3 had consistently high scores on tests of fluid intelligence; they scored above the mean on the Digit Span test, and two of them scored more than one standard deviation above the mean on the Digit Symbol and Ravens tests. Furthermore, they were more internal on both the generalized and the health-specific measures of locus of control. No significant differences were found among the wait ing list group when t-tests were performed between attend ers and total Time 3 drop-outs. However, it appeared that the drop-outs were less physically fit; none of the four of them had been able to complete the modified step-test at Time 1. These results suggest that drop-outs differed from at tenders on the initial levels of a number of variables. It is interesting to note that among the exercise group, these variables were primarily psychological; it does not appear that the physiological and cognitive variables distin guished between exercisers and drop-outs at pre-test. Hence, this suggests that the exercise drop-out was likely to hold different attitudes--such as more external per- 78 ceived locus of control and greater depression--but did not necessarily differ from the attender in physical fitness or in problem-solving capabilities. However, results for drop-outs from the control groups suggest that they may be characterized in a different way. In addition to indicating greater psychological distress, the control group drop-outs were less physically fit than the individuals who stayed in the program. Conversely, the control group drop-outs also indicated more internality than attenders and performed better than average on tests of fluid intelligence. Thus, control group drop-outs tend ed to be those functioning at either the high or the low end of the spectrum of psychological well-being, those with a greater than average capacity for fluid intelligence, and those who were less physically fit. Of further interest was the finding that only two of the drop-outs were married at Time 1, while the rest were either widowed, divorced, or never married. This finding fits with reports in the literature that marital status is one of the best predictors of exercise adherence (Gale et al., 1984). Of the two women who were married, one joined the program with her husband and they dropped out together, while the other dropped out because of marital difficulties which forced her to leave her home and move away from Los Angeles. Finally, the relationship between age and adherence 79 was examined. Correlations between age and the number of exercise days attended were not significant, and t-tests revealed no significant differences between mean age of participants at Times 1, 2 or 3. Thus, age was not a pre dictor of adherence to exercise among this sample. 80 CHAPTER V DISCUSSION This study was designed to improve on several aspects of past research in this area. While past research used institutionalized elderly subjects or only male subjects, or did not randomly assign subjects to conditions, the present study increased both internal and external validity by studying randomly assigned groups of community-residing older adults. Many past studies in the community have been devoted to studying extremely fit, high-SES older adults, while this study examined a racially-mixed, moderate-to-low income group. Thereby, generalizability was improved. The results of this study indicate several effects of exercise programs for the elderly, and thus expand upon past research in this area. Depressive symptomatology ap peared to decrease among the exercise group and anxiety was somewhat reduced which is similar to results found among younger populations (McCann & Holmes, 1984; Greist et al., 1979; Felkins, 1976). While the Morgan et al. (1970) study of a 6-week exercise intervention did not demonstrate an effect on depressive symptoms in the elderly, the present results suggest that an intervention of longer duration may facilitate a reduction of depressive symptoms. 81 In particular, females seemed especially to benefit from the exercise intervention. While this observation may reflect the fact that females began the program with higher levels of depressive symptomatology and less flexibility than males, it may also suggest that older women stand to gain as much as, if not more than, their male peers by tak ing part in an exercise program. The apparent utility of exercise programs for older females has not been demon strated in past studies that examined only males (Buccola & Stone, 1975) or that postulated a lack of motivation among females (Clark et al., 1975). In the present study it appeared that males and females with higher than average depressive symptoms responded to the program in different ways; males tended to drop out early while females contin ued to participate. While regression toward the mean was one explanation of the effects observed for depression, the sex differences noted here mitigate this as the sole ex planation for the findings. Correlations were found between improvement on various physiological indicators and increases on several psycho logical indicators, including both internal locus of con trol and mastery. This result corroborates the findings of Perri and Templer (1985) suggesting that greater internal ity among the aged was associated with participation in an aerobic exercise class. In addition, improvement on all five measures of cog- 82 nitive functioning was correlated significantly with im provement on various physiological indicators, particular ly cardiovascular measures. This elaborates upon Powell's (1974) finding that institutionalized aged improved on the Ravens test after an exercise program, and it would sub stantiate the possible contribution of exercise toward the older individual's capacity for fluid intelligence. In general, the physiological results were inconsis tent and problematic. While the blood pressure and heart rate measures did not indicate change among this popula tion, the measures of flexibility and weight did indicate change. Possibly blood pressure and heart rate were not adequate indicators of change due to the older age range of the sample and the relatively large proportion of sub jects taking medication for high blood pressure. This lack of responsiveness of blood pressure and heart rate has been reported by other researchers (e.g. Blumenthal et al., 1981) as has the failure to establish a training effect among exercise participants (Morgan et al., 1970). At the same time, there was little indication that the results for flexibility and weight were specific to exer cise. Indeed, it is possible that these effects resulted from participants' general commitment or motivation to im prove health. In short, the sample studied was not phys ically fit to begin with and the exercise did not appear to contribute substantially to overall physical fitness. 83 Most of the significant results of the study were de- rived from analyses with the control groups pooled. Thus, it is not possible to comment on the extent to which changes observed in the exercise group were the result of physical activity per se versus increased social interac- tion during the exercise classes. Nonetheless, in this study exercise appeared to lead to improvement in mood--especially less depressive symptom- atology--and to small physiological gains. Furthermore, physiological gains were related to improvement on mea- sures of fluid intelligence and, particularly among those with higher levels of participation, to mastery. The de- pression and mastery findings were further substantiated by informal comments made by participants. ~selected comments about the program included: "On the days that I exercise I'm a whole different person. Usually I just sit around the apartment, but on Monday, Wednesday, and Friday I come back and shower the sweat off and do things. I feel vital." "I leave class feeling so much better." "That's the reason I do exercise. know that I can do it." 'Cause I like to "I feel so much better after the exercises. You know, when you don't work anymore, you need some thing like this to keep you going." Clearly, these statements indicate improved mood and sense 84 of efficacy as a result of participation in the exercise program. Limitations of the Present Study Many of the improvements made in this study such as random assignment of subjects to conditions, use of a social control group, and studying non-institutionalized elderly, posed hurdles for the execution of the study. In deed, this project was plagued by a number of the dilemmas and barriers encountered by any field research with an ex perimental design. First, random assignment of subjects to conditions appeared to lead to drop-out, if participants were assigned to a condition that was not their first choice. In this project, despite great efforts to make the social activities program attractive, exercise was pre ferred by the vast majority. Second, recruitment of sub jects itself was made somewhat awkward by the need for ran dom assignment, since the advertising could not specify the condition to which a recruit would be assigned. Third, the experimental design required a pre-test which precluded studying ongoing groups and necessitated the recruitment of those not currently participating in the activity under study. Participants' lack of familiarity with not only the program but the instructors and the scheduling of classes appeared to be a further barrier to recruiting subjects and to maintaining them in the program. 85 A further experimental requirement was that the same intervention be offered to all subjects in each condition. In the exercise classes, an attempt was made to offer a standardized group experience, while encouraging partici pants to pace themselves. Since subjects were relatively inexperienced with exercise programs the intervention was necessarily less vigorous than the kind of program that might be conducted for older adults exercising at a gym or health club. Thus, it was more difficult to assure a training effect, despite the length (12 weeks) and inten sity (3 times per week for one hour) of the program. Hence, in conducting this research project, the decision was made to maintain rigorously the experimental design and methodology. The price of internal validity appeared to be some loss of external validity. Drop-Out Apart from the contribution of the experimental design to difficulties in executing this study, drop-out during the study itself posed further problems. While drop-out from the exercise group was not particularly severe, ad herence among the social control group was less than sat isfactory. By the end of the first 12 weeks there was 27% drop-out among this group and attendance for those remain ing ranged from 10% to 94%. Only 27% attended 3/4 of these sessions. As a consequence of the limitations of this so- 86 cial contact group, results of this study were derived mainly from the pooled control groups. At the same time, these results suggest that drop-out is a problem not limited to exercise. Drop-out occurred in both the social activity and waiting list control groups. With such a high drop-out rate in the social control group, the present results suggest that this kind of group did not serve as an adequate control in this study and that its use in future studies of this type is questionable. Reasons given for non-compliance--either missing classes or dropping out altogether--were obtained informal ly by questioning those who could be located or their friends who remained in the program. Ill health predomi nated as a reason for dropping out, either medical problems of one's own or responding to the declining health of a loved one. Another reason for missed classes was competing commitments, such as waiting for a television repairman to arrive. If pressed for explanations, some drop-outs cited personality clashes, in particular, higher functioning par ticipants became impatient with those they perceived as functioning below their level. Thus, the reasons given for drop-out tended to be personal rather than attributes of the program. However, exploration of reasons for drop-out or non-compliance in exercise programs designed for older adults is an area in need of further systematic investiga tion. 87 An underlying problem resulting from drop-out as well as from the requirements of the experimental design was the low N relative to the number of subjects that had been an ticipated. While this was problematic for the study be cause statistical power was reduced, it is a problem not uncommon to the research in this area (e.g. Blumenthal et al., 1981; Clark et al., 1975). Furthermore, decreased at tendance among those still participating in the program may have reduced the effect of the intervention. Future studies in this area must address the need to increase the number of participants and maintain high attendance levels without endangering the external validity of the research. Measures The psychological measures used in this study appeared to provide a reliable indicator of psychological well-be ing. Correlations among the three locus of control scales were significant as were correlations among the measures of depression, anxiety, and self-rated happiness. Likewise, the cognitive measures appeared to provide a reliable mea sure of fluid intelligence, though the writing numbers test was perhaps least useful as an indicator of the construct due to the extremely large variance on the measure. The physiological measures as a group proved to be less reliable. The variances for the balance and arm strength measures were extremely large and rendered these 88 measures impractical for use. Likewise, the standing flex ibility measure appeared to be unreliable. On the other hand, the heart rate and blood pressure measures formed two separate but reliable indicators of physiological status. In particular, resting and final diastolic blood pressure seemed to be the strongest indicators of the construct. Though the small size of effects for the psychological measures may have resulted from reduced power due to low N, a larger sample probably would not have been sufficient for improving the reliability of the physiological mea sures. Rather, for these measures it would seem that more precise instruments were needed for measuring changes--in strength as well as in cardiovascular conditioning--among this population. Other than the measure of perceived physical exercise, which served primarily to confirm subjects' increased ac tivity level as a result of the program, the health behav ior measures did not indicate any consistent shift in health behaviors as a result of the intervention. Theoretical Implications The lack of significant change among the exercise group on many of the psychological variables is a finding not inconsistent with past research in this area (Blumen thal et al., 1981). However, it is of interest that sig nificant correlations were found between improvement on 89 various physiological indicators and improvement on several psychological measures including the mastery scale. This result would suggest that the exercise and the exercise ef fect are an important component of improved psychological well-being. In addition, this finding contributes to our theoretical understanding of locus of control. Because physiological change correlated with change on the measure of generalized locus of control, the present results may lend support to the notion that locus of control is poten tially malleable among an older population. Similarly, for the cognitive measures, though the ex ercisers as a group did not improve significantly, those who did show physiological improvement were also more like ly to improve on measures of fluid intelligence. Thus, it would appear that participation in the exercise program contributed to greater psychological well-being and cogni tive functioning among those who attended and showed phys iological gains. However, attending the exercise class without physiological improvement also may have been as sociated with improvements on the psychological indices. Thus, it may be especially important to explore further the mechanisms of change resulting from participation in exer cise. While physiological improvement would appear to ex plain increases in perceived control and mastery, it may not explain decreases in levels of depression. Indeed, change on psychological or cognitive measures may be re- 90 lated to the release of endorphins or enkephalins in the central nervous system, as researchers have suggested (Emrich, 1982; Markoff, Ryan & Young, 1982), or there may be yet other physiological or psychosocial components of exercise that are responsible for observed changes. Practical Implications Participants in the exercise program not only enjoyed themselves, but they improved on indicators of psychologi cal and physiological well-being. Plainly, there are older adults in the community who enjoy exercise programs and who stand to benefit from such programs. At the same time, there are large numbers of individuals working with com munity-residing elderly who desire information about exer cise programs for older adults. While there is widespread social endorsement of exer cise for people of all ages, there is also a dearth of in formation readily available to the public concerning exer cise for older adults. This gap between social attitudes and available information leads not only to confusion about the value of exercise programs, but also to potentially harmful consequences for those older individuals residing in the community who wish to participate in exercise classes. Programs advertised for older adults may include exercises inappropriate for this age group and, by the same token, older adults may have difficulty selecting a pro- 91 gram suited to their needs. The exercise program conducted for this study included exercises that took into account morphologic changes com mon among older adults. Furthermore, an attempt was made to include exercises that would be useful to the partici pant and that could be practiced at home, such as the rapid walking and the sitting-to-standing exercise. This avoided the common problem of providing the older adult with exer cises that could be performed only in a gym or a laboratory setting with special equipment. In order for older adults to benefit from exercise programs, the practicality and utility of the exercises should be a factor in program development. Moreover, in formation about exercise programs for the elderly must be disseminated not only to those planning or conducting exercise classes, but also to older adults and to their physicians who may be prescribing exercise. This informa tion should encompass not only the positive aspects of ex ercise for the older adult--such as improved psychological well-being and weight loss--but also the caveats and limi tations of exercise for older adults. The present study suggests that exercise does not universally benefit older adults, that its impact on participants may differ between individuals, and that exercise, therefore, should not be liberally prescribed as a panacea for older adults. In stead, it is important to understand more about the effect 92 of exercise upon older adults which entails first learning more about group differences, such as gender and ethnic status differences, that may affect the older adult's re sponse to exercise. Second, and of equal importance, indi vidual differences in exercise response between and among older adults must be more systematically explored. It is possible that the intervention in this study was too conservative, i.e. the exercise was not strenuous enough to produce the hypothesized physiological and psy chological gains. More generally, aerobics may not have been an adequate intervention for producing a training ef fect. In order to produce significant cardiovascular changes, it may be necessary to employ a more rigorous training program including activities such as bicycle er gometer workouts and weight lifting. These are important practical considerations which also have a bearing on future research in this area. Research .!!!!rlications One suggestion for future research in this area would be to continue studying and expand upon the kinds of exer cises that older adults have readily available to them, rather than studying exercise that is available only in a particular setting. Furthermore, future studies might at tempt to match the exercise not only to the needs of this age group but also, within this group, more closely match 93 exercise to the individual needs of the older person. This strategy would serve to maximize physiological effects, and thereby provide a more precise testing ground for the ef fects of exercise on the individual's psychological well being, assessed with a variety of measures such as per ceived mastery, depression, and anxiety. Likewise, in a more individualized program the apparent relationship be tween physiological and cognitive improvement might be teased out. While the size of the results in the present study was consistently small, the overall pattern of results con firms the types of effects hypothesized. The small size of the effects probably results, in part, from two different factors which future research might address. First, for the psychological measures, the small N represents reduced power. Second, for the physiological measures the neces sity of conducting measures with inexpensive and portable equipment limited sensitivity in an area where only small changes could be expected. Future research would benefit from studying larger groups, if possible, and using more sophisticated means of measuring physiological change, while maintaining ecological validity, (i.e. studying com munity-residing older adults,) as well as substantive val idity of psychological measures, (i.e. examining proven constructs relevant to the well-being of older adults.) Though the present results suggest that exercise may 94 not universally benefit adults, the prospects for future research in this area are multitudinous, and the implica tions for the health and well-being of older adults resid ing in the community are far-reaching. As evidenced by the numerous requests for information regarding this project, there is a need for practical as well as theoretical infor mation in this field. Both older adults and the profes sionals who work with them seek information regarding plan ning and implementation of programs. By continuing to pur sue the kinds of questions addressed in this project and similar studies, not only will the theoretical understand ing of relationships between physiological, psychological, and cognitive functioning be enhanced, but the practical needs of older adults residing in the community will more closely be met. 95 REFERENCES American Alliance for Health, Physical Education and Recre ation. (1979). AAHPERD Health Related Physical Fitness Test Manual. WashJ.ngton, D.C.: AAHPERD. Barry,A., Steinmetz,J., Page,H. & Rodahl,K. (1966). The Effects of Physical Conditioning on Older Individuals. II. Motor Performance and Cognitive Function. Journal of Gerontology, 21, 192-199. Birren,J.E. & Botwinick,J. (1951). The Relation of Writing Speed to Age and to the Senile Psychoses. Journal of Consulting Psychology, 15, 243-249. Blazer,D. (1980). The Epidemiology of Mental Illness in Late Life. In E.W. Busse & D.G. Blazer (eds.), Handbook of Geriatric Psychiatry. New York: Van Nostrand ReJ.n nold. Blumenthal,J., Schocken,D., Needels,T. & Hindle,P. (1982). Psychological and Physiological Effects of Physical Con ditioning on the Elderly. Journal of Psychosomatic Re search, 26(5), 505-510. Bradburn,N.M. (1969). The Structure of Psychological Well Being. Chicago: Aldine. Brim,O.G.,Jr. (1974). The Sense of Personal Control Over One's Life. Invited address at the annual convention of the American Psychological Assocation. Buccola,V.A. & Stone,W.J. (1975). Effects of Jogging and Cycling Programs on Physiological and Personality Var iables in Aged Men. Research Quarterly, 46(2), 134-139. Carter,R. (1977). Exercise and Happiness. Journal of Sports Medicine, 17, 307-313. Clark ,B.A., Wade ,M.G. 1 Massey, B. H. & Van Dyke ,R. ( 197 5) • Response of Institutionalized Geriatric Mental Patients to 12-1-Veek Program of Regular Physical Activity. Journal of Gerontology, 30(5), 565-573. 96 Dishman,R.K. (1982). Compliance/Adherence in Health-Related Exercise. Health Psychology, 1 (3), 237-267. Duke,M., Johnson,T.C. & Nowicki,S.,Jr. (1977). Effects of Sports Fitness Camp Experience on Locus of Control Orientation in Children, Ages 6 to 14. Research Quarter ly, 48(2), 280-283. Duke University Center for the Study of Aging and Human Development. (1978). Multidimensional Functional Assess ment: The OARS Methodology, A Manual (2nd ed.), Durham, N.C.: Center for the Study of Ag1ng and Human Develop ment. Elsayed,M., Ismail,A.H. & Young,R.J. (1980). Intellectual Differences of Adult Men Related to Age and Physical Fitness Before and After an Exercise Program. Journal of Gerontology, 35(3), 383-387. Emrich,H. (1982). A Possible Role of Opioid Substances in Depression. Advances in Biochemical Psychopharmacology, 32, 77-84. Folkins,C.H. (1976). Effects of Physical Training on Mood. Journal of Clinical Psychology, 32(2), 385-388. Folkins,C.H. & Sime,W.E. (1981). Physical Fitness Training and Mental Health. American Psychologist, 36(4), 373- 389. Frankel,L. (1977). Be Alive As Long As You Live. New York: Harper~ Row, Inc.---- Gale,J.B., Eckhoff,~7.T., Mogel,S.F. & Rodnick,J.E. (1984). Factors Related to Adherence to an Exercise Program for Healthy Adults. Medicine and Science in Sports and Exer cise, 16(6), 544-549. Gallagher,D. & Thompson,L. (1981). Depression in the Elder ~ A Behavioral Treatment Manual. Los Angeles: Univer sity of Southern California Press. Gatz,M., Siegler,I., George,L. & Tyler,F. (1985). Attribu tional Components of Locus of Control: Longitudinal, Retrospective, and Contemporaneous Analyses. In M.M. Baltes & P.B. Baltes (eds.), Aging and the Psychology of Control. Hillsdale, N.J.: Erlbaum. 97 Greist,J.H., Klein,M.H., Eischens,R.R., Faris,J., Gurman,A. & Morgan,W.P. (1979). Running as Treatment for Depres sion. Comprehensive Psychiatry, 20(1), 41-54. Harris,R. & Frankel,L.J., eds. (1977). Guide to Fitness Af ter Fifty. New York: Plenum Press. Himmelfarb,s. & Murrell,S. (1983). Reliability and Validity of Five Mental Health Scales in Older Persons. Journal of Gerontology, 38(3), 333-339. Horn,J.L. & Cattell,R.B. (1966). Refinement and Test of the Theory of Fluid and Crystallized Intelligence. Journal of Educational Psychology, 57, 253-270. Horn,J.L. & Donaldson,G. (1976). On the Myth of Intellec tual Decline in Adulthood. American Psychologist, 31, 701-719. Kay,D.W.K. & Bergmann,K. (1980). Epidemiology of Mental Disorders Among the Aged in the Community. In J.E. Bir ren & R.B. Sloane, eds., Handbook of Mental Health and Aging. Englewood Cliffs, N.J.: Prentlce-Hall. Kuypers,J.A. (1972). Internal-External Locus of Control, Ego Functioning, and Personality Characteristics in Old Age. The Gerontologist, 12, 168-173. Kuypers,J.A. & Bengtson,V.L. (1973). Social Breakdown and Competence. Human Development, 16, 181-201. Lachman,M., Baltes,P., Nesselroade,J. & Willis,s. (1982). Examination of Personality-Ability Relationships in the Elderly: The Role of Contextual (Interface) Assessment Mode. Journal of Research in Personality, 16, 485-501. Langer,E., Rodin,J., Beck,P., Weinman,C. & Spitzer,L. (1979). Environmental Determinants of Memory Improve ment in Late Adulthood. Journal of Personality and Social Psychology, 37: 2003-2013. Lau,R.R. & Ware,J.F. (1981). Refinements in the Measure of Health-Specific Locus-of-Control Beliefs. Medical Care, 19(11), 1147-1158. Lawton,M.P. (1977}. Morale: What Are We Measuring? In C.N. Nydegger, ed., Measuring Morale: A Guide to Effective Assessment. Nashington,D.c.:·Gerontologlcal Soc1ety. 98 [ . .-···· Lewinsohn,P.M., Munoz,R.F., Youngren,M.A. & Zeiss,A.M. (1978}. Control Your Depression. Englewood Cliffs, N.J.: Prentice-Hall. ---- Markoff,R., Ryan,P. & Young,T. (1982}. Endorphins and Mood Changes in Long-Distance Running. Medicine and Science in Sports and Exercise, 14(1}, 1-15. McCann,I.L. & Holmes,D.S. (1984}. Influence of Aerobic Ex ercise on Depression. Journal of Personality and Social Psychology, 46(5}, 1142-1147.- --- Minister of State: Fitness and Amateur Sport (Canada}. (1979}. Standardized Test of Fitness: Operations Manual. Moos,R.H., Cronkite,R.C., Billings,A.G. & Finney,J.W. (1983). Health and Daily Living Form Manual. Social Ecology Laboratory, Veterans Adm1n1strat1on and Stanford University Medical Center. Morgan,W.P. (1974}. Exercise and Mental Disorders. In A.J. Ryan & F.L. Allman,Jr., eds., Sports Medicine. New York: Academic Press. Morgan,W.P., Roberts,J.A., Brand,F.R. & Feinerman,A.D. (1970). Psychological Effect of Chronic Physical Activ ity. Medicine and Science in Sports, 2(4}, 213-217. Palmore,E. & Luikart,C. (1972}. Health and Social Factors Related to Life Satisfaction. Journal of Health and Social Behavior, 13, 68-79. Pearlin,L.I. & Schooler,C. (1978}. The Structure of Coping. Journal of Health and Social Behavior, 19, 2-21. Perri,S. & Templer,D. (1985). The Effects of an Aerobic Ex ercise Program on Psychological Variables in Older Adults. International Journal of Aging and Human Devel opment, 20(3), 167-172. Powell,R.R. (1975). Effects of Exercise on Mental Function ing. Journal of Sports Medicine, 15, 125-129. Powell,R.R. (1974). Psychological Effects of Exercise Ther apy Upon Institutionalized Geriatric Mental Patients. Journal of Gerontology, 29(2), 157-161. Radloff,L. (1977). The CES-D Scale: A Self-Report Depres sion Scale for Research in the General Population. Ap- plied Psychological Measurement, 1, 385-401. - 99 Raven,J.C. (1963). Guide to Using the Coloured Progressive Matrices. London: H.K. Lewis & Co., Ltd. Rodin,J. (1980). Managing the Stress of Aging: The Role of Control and Coping. In H. Ursin & s. Levine, eds., Cop ing and Health. New York: Academic Press. Rotter,J. (1975). Some Problems and Misconceptions Related to the Construct of Internal Versus External Control of Reinforcement. Journal of Consulting and Clinical Psychology, 43, .56-67. - Rotter,J.B. (1966). Generalized Expectancies for Internal Versus External Control of Reinforcement. Psychological Monographs, 80, ( 1, Whole No. 609) • Sager,K. {1984). Exercises to Activate Seniors. The Physician and Sportsmedicine, 12(5), 144-151. Stamford,B.A., Hambacher,W. & Fallica,A. {1974). Effects of Daily Physical Exercise on the Psychiatric State of In stitutionalized Geriatric Mental Patients. Research Quarterly, 45(1), 34-41. Tredway, V .A. { 1978) • Mood and Exercise in Older Adul,ts. Dissertation Abstracts, 2531-B. u.s. Department of Health,Education, and Welfare. (1978). Health-United States-1978. DHEW Publication No. (PHS) 78-1232. Wechsler,D. (1981). WAIS-R Manual. New York: Harcourt Brace Jovanovich. 100 APPENDIX Measures 1. Physiological Data Sheet 2. Cognitive a. Order of tests b. Record sheet for copying numbers c. Record sheet for copying words 3. Self-Administered Questionnaire 101 !'hvsiological !) Handedness: Which hand do you use to write with? Left Right II) Weight III! Flexibility: A) Sit and reach: (NB:If subject can not sit on floor, please do this on table.) B) Side bend: Left side Right side __ _ IV) Balance: A)Length of time liftina right foot B)Length of ~ime lifting ~eft foot Vl Strength: A) Right arm, length of time Bl Left arm, length of time VI) Resting Heart Rate Resting B~ood Pressure S~s Dia (~B: If resting HR is ov2' 100 or resting BP is over 170/90, do not procede with step test.) VII) Step ':'est: Stage 1: HR (50 bpml Stage 2: HP. ( 60 bpm) Stage 3: HR ( 80 bpm) Final Blood Pressure Sys Dia __ _ NB: If subject is unable to perform step-test using two steps, complete the test using only ~ step. 102 Co::nitive Order of tests: 1. Couving (page of numbers) Record total time to complete the task in seconds at bottom of sheet. (Remind subjects to work left to right on each line.) 2. \\ore; Flue'l£J: (page of words) Stop after 2 minutes. (Again, subject should be working left to right on each line.) 3. Digit Sv~bol Stop after 90 seconds. 4.~~ 5. Raven's Progressive Matrices 103 _g 8 _g 4 8 4 _g 1 4 1 4 0 1 l 1 - • _g l 8 2 1 8 l 1 l 2 .2 1 _g 8 1 l 6 1 8 1 4 1 1 2 6 1 6 1 6 8 8 2 1 6 4 8 1 1 8 8 2 1 l l Q 1 2 ..... 1 1 8 l l 4 1 4 1 6 104 area manage protest today precedent committee wcrk exaggerate feature :alled figure mother swirl private already southern serve maternity where pile shatter grou::::! began estimate hungr:: river village horse between farm s-,:read beside quart drift relief inch illness emergency quiet Marines have midl''light tallest cancel postpone reserve score wetted arrest high school struck buf:'et pass disaster rough sta:r rescue thin dome situation open exept snow a anger previous Nevada woman minute northern 105 Information Please complete the following background information. 1. How old are you? __ _ 2. What is your birthday? ___ __,..,--...,.--- month,day year J. Sex Male Female 4. What is your present marital status? Married ---Divorced ---Se'!'o.rated ---.N'idowed :::::single 5. What ethnic background do you consider yourself to be? Caucasian --Black --Hispanic (specify) _____ _ --Asian(specify) :::::other(specify) _______ _ 6. What is your highest level of education? Number of y~s comnleted ---High school graduate ---Some college, but no degree ---B.A./B.S. ---Some graduate work, but no degree -M.A. --Ph.D. :==other professional degree(specify) _____ _ 7, What is your religious preference? Catholic ---Protestant ---Jewish ---Buddhist ---Moslem :::::o~er(specify) _________ __ ___ None -1- 106 -2- 8. Which of the following categories approximates your income !or last year? · 1 0-$1,999 -- 2, OOO-$J, 999 ----- 4,000-~5.999 --- 6,000- 7.999 ::::: 8,000- 9.999 --~10 '000-:14' 999 15,000- 19,999 -- 20,000- 24,999 ::::: 25,000 or more 9. What kind of work have you done most of your life? (Place an 'x' beside the one category that describes what you did most of your lif~ urofessional or higher executive (e.g. bank or large ---business vice-president, doctor, professor, government official) manager of business, other professional (e.g. personnel -------manager, accountant, nurse, pharmacist, sheriff, teacher) administrative personnel, protective services (e.g. insurance ------adjuster, credit manager, private secretary, policeman, fireman, armed forces) technicians, clerical and sales workers (e.g. sales rep., ------insurance agent, bank teller, dental technician, secretary) skilled manual and service workers (e.g. carpenter, chef, -------auto mechanic, post-man, landscape gardener, repairman) semi-skilled (e.g. equipment operator, bus driver, bar tender, ------coal miner, hairdresser, waitress, cook) laborer (e.g. janitor, stagehand, farm worker, construction ------worker, cafeteria worker, domestic) ______ housewife, homemaker 107 -3- Please indicate the appropriate response category for each of the following statements. l. 2. < __.. 4. "I drink 1 Never "I smoke 1 Never "I engage and doing 1 Never alcohol (beer, wine, or hard liquor). n 2 3 Rarely Occasionally cigarettgs, cigars, or pipes. " 2 3 Rarely Occasionally i~ physical exercise (other than running housework) • " 2 Rarely 3 Occasionally "I feel that my 1 diet is well-balanced." 2 3 Strongly Disagree Mildly Mildly Disagree Agree S. "I feel that I sleep well." 1 Strongly Disagree 2 Mildly Disagree 3 Mildly Agree 6. In general, how would you rate your health? Excellent Good -- Fair. ____ _ Poor ____ _ ?. How would you say you are these days? Very happy Pretty hap~p~y~- Not too happy 4 Frequently 4 Frequently errands 4 Frequently 4 Strongly Agree 4 Strongly Agree 108 -4- For each of the following statements please ~ ths number beneath it ~JilQ..U c:.losely describes your opinion. ~ St3y~ng well has little or nothing to do with chance. 1 STRO~GLY DISAGREE 2 3 4 5 6 7 STRONGLY AGREE 2 Seeing a doctor for regular check-ups is a key factor in staying healthy. 1 STRONGLY DISAGREE 2 3 4 5 6 7 STRONGLY AGREE 3 Doctors can rarely do very much for people who are sick. 1 STRONGLY DISAGREE 2 3 4 5 6 7 STRONGLY AGREE 4 Anyone can learn a few basic health principles that can go a long way in preventing illness. 1 STRO~GLY DISAGREE 2 3 4 5 6 7 STRONGLY AGREE 5 People's ill health results from their own carelessness. 1 STRONGLY DISAGREE 2 3 4 5 6 7 STRONGLY AGREE 6 Doctors relieve or cure only a few of the medical problems their patients have. 1 STRONGLY DISAGREE 2 3 4 5 6 7 STRONGLY AGREE 7 There is little one can do to prevent illness. 1 STRONGLY DISAGREE 2 3 4 5 6 7 STRONGLY AGREE 109 a No matter what anybody does, there are many diseases that can just wipe you out. 1 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE 9 I have a lot of confidence in my ability to cure myself once I get sick. 1 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE -5- 10 Whether or not people get well is often a matter of chance. 1 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE 11 People who never get sick are just plain lucky. 1 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE 12 Doctors can almost always help their patients feel better. l 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE u· The seriousness of many diseases is overstated. l 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE 14 When it comes to health, there is no such thing as "bad luck. .. l 2 3 4 5 6 7 S:'RONGLY STRONGLY DISAGREE AGREE 110 ~5 In the long run, people who take very good care of themselves stay healthy and get well quick. 1 2 3 4 5 6 7 STRONGLY STRONGLY DIS.li,.GREE AGREE 16 Recovery from illness requires good medical care more than anything else. 1 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE 17 In today's world, few diseases are totally debilitating (crippling). l STRONGLY DISAGREE 2 3 4 5 6 7 STRONGLY AGREE 18 Recovery from illness has nothing to do with luck. l STRONGLY DISAGR.EI! 2 3 4 5 6 7 STRONGLY AGREE 19 Most people are helped a great deal when they go to a doctor. l STRONGLY DISAGREE 2 3 4 5 6 7 STRONGLY AGREE 20 There are a lot of medical problems that can be very serious or even fatal (can kill you) . l 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE 21 Healthwise, there isn't much you can do for yourself when you get sick. 1 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE -6- 111 -7- 22 Doctors can do very little to prevent illness. 1 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE 23 "Taking care of yourself" has little or no relation to whether you get sick. 1 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE . AGREE 24 Some kinds of illness are so bad that nothing can be done about them. 1 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE 25 If I get sick, it's usually my own fault. 1 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE 26 Many times doctors do not help their patients to get well. 1 2 3 4 5 6 7 STRONGLY STRONGLY DISAGREE AGREE 27 Good health is largely a matter of fortt:.ne. 1 2 3 4 5 6 7 S':'RONGLY STRONGLY DISAGREE AGREE 112 -8- Place an 'x' in the square below the statement which best describes how often you felt this way DURING THE PAST WEEK. D~RING THE PAST WEEK: a. b. c. d. e. f. g. h. j. k. 1 m. n. o. p. q. r. s. t. u. I was bothered by things that don't usually bother me . . . I did not feel like eating; my appetite was poor . . . I felt that I could not shake off the blues even with help from my family or friends . I felt that I was just as good as other people . . . . . . I had trouble keeping my mind on what I was doing. . . I felt depressed . . . . I felt that everything I did was an effort . . . . . . . . I felt hopeful about the future. I thought my life had been a failure . . . . . . . . I felt fearful . . . . . . My sleep was restless . . . I was ha;;py . . . . . . I talked less than usual. . . . I felt lonely . . . . . People were unfriendly . . . I enjoyed life . . . . . . . . I had crying spells . . . . . I felt sad . . . . . . I felt that people disliked me . I could not get 'going' . I felt anxious or tense . . . Rarely or Some of Occasion- none of the time ally the time Most of the time 113 Rotter Scale This is a questionnaire to find out the way in which certain important events in our society affect different people. Each item consists of a pair of alternatives lettered a or b. Please select the one statement of each pair (and only one) which you more stron~ly believe to be the case as far as you're concerned. Be sure to select the one you actually believe to be.more true rather than the one you thi~k you should choose or the one you would like to be true. This is a measure of personal belief: obviously there are no ri~ht or wron! answers. Your answers to the items on this inventory are to be recorded by circling alternative (A or B) on the answer sheet, in front of the statement which you believe to be more true for each item. Please answer these items carefully but do not spend too muc~ time on any one item. Be sure to fina an answer for every choice. In some instances you may discover that you believe both state ments or neither one. In such cases, be sure to select the one you more stron~ly believe to be the case as far as you are concerned. Also try to respond to each item inde~endentlv when making you choice; do not be influenced by your previous choices. l ~ stronglv believe !h!!: l) A. Many of the unhappy thin!S in people's lives are pertly due to bad luck. B. People's misfortunes result from the mistakes they make. 2) A. One of the major reasons why we h&ve wars is because people don't take enou!h interest 1n politics. B. There will always be wars, no matter how hard people try to prevent them. 3) A. In the long run people !et the respect they deeerve 1n this world. B. Unfortunately, an individual's worth often passes unreco!n!zed no matter how hard he tries. h) A. The idea that teachers are unfair to students is nonsense. B. Most students don't realize the extent to which their !rades are influenced by accidental happenin~s. 5) a. Without the right breaks one cannot be an effective leader. B. Capable people who fail to become leaders have not taken advantage of their opportunities. 114 -10- 1 !£!! stron!lY believe ~: 6) A. No matter how hard you try some people just don't like you. B. People who can't !et others to like them, don't understand how to get along with others. 7) A. I have often found that what is going to happen will happen. B. Trustin! to rate has neTer turned out aa well for •e as makiD! a decision to take a defin1 te course of action •. 8) A. In the case or the well prepared student there is rarely i~ ever such a thing as an unfair test. B. Many times exam questions tend to be so unrelated to course workf that s~udyin! is really useless. 9) A. Becoming a success is a matter of hard work, luck has little or nothing to do with it. B. Getting a !OOd Job depends mainly on bein! in the right place at the ri!ht time. 10) A. The average citizen can have an influence in !OVernment decisior.. B. This world is run by the few people in power, and there is not much the little !UY can do about it. ll) A. When I make plans, I am almost certain that I can make them work. B. It is not always wise to plan too far ahead because many thin!s turn out to be a matter of !OOd or bad fortune anyhow. 12) A. In my case getting what I want has little or nothing to do with luc.k. B. Many times we mlgnt ju~t as well decide what to do by fl~p p1n! a coin. 13) A. Who !e~d to be the boss often depends on who was lucky enough to be in the ri!nt place first. B. Getting people to do the ri!ht thing depends upon ability, luck has little or nothing to do with it. 14) A. As far as world affairs are concerned, most of us are the victims of forces we can neither understand, nor control. B. By taking an active part in political and social &~fairs the people can control world events. 115 -11- 1 !2!! stron!li believe that: 15) A. Most people don't realize the extent to which their lives are controlled by accidental happenin!S• B. There really is no such thin! as "luck". 16) A. It is hard to know whether or not a person really likes you. B. How many friends you have depends upon how nic' a person you are. 17) A. With enou!h effort we can wipe out political corruption. B. It 1s difficult for people to have much control over the th:ngs politicians do in office. 18) A. Sometimes I can't understand how teachers arrive at the !rades they give. B. There is a direct connection between how hard I study and the !;rades I get. 19) A. Many times I feel that I have little influence over the things that happen to me. B. It is impossible tor me to believe that chance or luck plays an important role in my life. 20) A. People are lonely because they don't try to be friendly. E. There's not much use in tryin! too hard to please people, if they like you, they like you. 21) A. What happens to me is my own doing. B. Sometl~es I feel that I don't have enou~h control over the direction my life is taking. - 22) A. rn the lon! run the bad things that happen to us are balanced by the good ones. B. Most misfortunes are the result or lack of ability, i!norance, laziness, or all three. 23) A. Most of the time I can't understand why politicians behave the way they do. B. In the lon! run the people are responsible for bad government on a national as well as on a local level. 116 -12- Pearlin Scale Here are seven statements. Each one represents feelings or attitudes that people often have. Tell me how much you personally agree or disagree with each one of them. 1. There is really no way I can solve some of the problems I have. 2. Sometimes I feel that I'm being pushed around in life. 3. I have little control over the things that happen to me. 4. I can do just about anything I really set my mind to do. 5. I often feel helpless in dealing with the problems of life. 6. What happens to me in the future depends mostly on me. 7. There is little I can do to change many of the important things in my life. A Strongly agree B Somewhat agree C Somewhat disagree D Strongly disagree 117 -13- Current Activity 1. In their recreation, leisure activities, or work, some people spend a lot of time in strenuous activity - like jogging, or running, playing handball or tennis, vigorous swimming, climbing, hiking, or doing heavy work around the house. Other people don't engage in this kind of strenuous activity at all. About how many hours do you spend, in an average week, in strenuous activities like these? (Circle one) None, don't do strenuous activity . . . . . . 0 1 hour or less ............. 1 2 to 3 hours a week 2 4 to 5 hours a week 3 6 to 10 hours a week . . . . . . . . . . . . . . . 4 More than 10 hours a week . . . . . . . . . . . . 5 2. Then there are activities that require a medium or moderate amount of physical activity - like dancing, playing golf, gardening, or working with home tools. About how many hours do you spend, in an average week, in medium or moderare-activities like these? (Circle one) None, don't do medium activity . . . . . . . 0 2 hours or less . . 1 3 to 5 hours a week .. 2 6 to 10 hours a week . . . . . . . . . . . . . . . 3 11 to 15 hours a week . . 4 More than 15 hours a week 5 118 -14- 3. Which one of these statements best describes your physical activity, in general? Not very active physically, usually only sitting or walking. Fairly active physically, moderate or strenuous (Circle one) . . 1 activity several times a week . . . . . . 2 Quite active physically, at least moderate activity every day . . . . . . . . . . . . . . 3 Extermely active physically, strenuous activity most days ........... 4 119 Activities Scale Please rate each activity listed below, accord1ne; to how much you enjoy participating in this activity, usin! the following scale: 1 2 3 4 2 not at ail moaerately very much l. Going for a leisurely walk. Rating: 2. Talking with friends. Rating: 3. Play in! cards. Rat in!: 4. Goifl! to the movies. Rat in!: c; .,. 6. 7. 1 1 l 8. 9. o. 1. 2. Dancing. Rating: Reading literature. Rating: Cooking. Rating: Solving a problem, puzzle, Ratin!;: crossword. Making things w1 th your Ratifl!: hands, such as clay objects. Shopping for new clothes. Rating: Planning or oresaniz1ne; Rating: something. Meditating. Rating: -15- How orten have these events occurred in your life in the past month: O•Not at all 1•1-6 times .2•7 times or more Frequency: Frequency: Frequency: Frequency: Frequency: Frequency: Frequency: Frequency: Frequency: Frequency: Frequency: Frequency: 120

Asset Metadata

Creator Emery, Charles Fiske (author)

Core Title The effect of physical exercise on cognitive and psychological functioning in community aged

Contributor Digitized by Interlibrary Loan Department (provenance)

Degree Doctor of Philosophy

Degree Program Psychology

Defense Date 12/01/1985

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

Tag OAI-PMH Harvest

Format application/pdf (imt)

Language English

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-493960

Unique identifier UC11297920

Identifier etd-Emery-579216.pdf (filename),usctheses-c3-493960 (legacy record id)

Legacy Identifier etd-Emery-579216.pdf

Dmrecord 493960

Document Type Dissertation

Format application/pdf (imt)

Rights Emery, Charles Fiske

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

Linked assets

University of Southern California Dissertations and Theses