The purpose of this study was to assess the relative influence of stimulus cues and response tendencies on the behavior of average and consistently or inconsistently overweight individuals. The female undergraduate volunteers who participated in the study were assigned to one of three weight groups on the basis of weight history, present weight, and triceps skinfold measurements. The three groups included consistently average, consistently overweight, and "fluctuator" subjects. This last group consisted of subjects whose weights over the past two years had varied between the average and overweight classifications. There were 20 subjects in each of the three groups. All subjects completed two experimental tasks and were also administered the Eysenck Personality Inventory (EPI), Form A. The first experimental task, used previously by Sikes, involved guessing the colors (black or red) of 120 consecutively presented cards. Seventy-five per cent of the first 90 cards in the series were black, while all of the remaining 30 cards were red. As expected, there were no performance differences between groups on the first 90 cards; however, on the last 30 cards consistently overweight participants made significantly more errors than either average or fluctuator subjects. This finding is consistent with Singh's deficit-in-response-inhibition hypothesis which maintains that overweight people have greater difficulty in changing established response tendencies than do people of average weight. It is noteworthy, however, that in the present study only people who had been consistently overweight for the past two years experienced more difficulty in changing their established responses. The second experimental task involved learning two paired associate word lists in an A-B/A-Br transfer paradigm. As expected, there were no performance differences between groups on the initial list. Contrary to expectations, however, there were also no differences between groups on the transfer list which required the suppression of previously established responses. Thus, in this situation consistently overweight subjects did not show the predicted deficit-in-response-inhibition effect. The results, then, offer partial support for Singh's interpretation of obesity in terms of differential response tendencies. No support is found for Schachter's interpretation which stresses the effects of external cues, since obese subjects did not show performance that was superior to that of average subjects at any point. Results from analyses of subjects' scores on the EPI indicated that there were no differences between the three weight groups on either extraversion or neuroticism. In addition to providing some support for Singh's hypothesis, the experimental findings in this study indicate that it is important to consider recent weight history as well as present weight when investigating behavioral differences between overweight and normal individuals. Implications of this research for treatment of overweight individuals were discussed. / Arts, Faculty of / Psychology, Department of / Graduate

The primary purpose of the present study was to examine the larger issue of color and emotional responses and, within that framework, to explore ways of specifying complex color displays. Several steps were involved in this investigation. First, a total of 80 color displays representing five levels of hue, two levels of value, two levels of chroma and four levels of motif were constructed. These were unique in that they accurately and systematically sampled the Munsell color space, and in the fact that they contained large numbers of color elements which were colorimetrically specifiable and which at the same time were arranged in such a way that they resembled color pictures. They thus bridged the gap between stimuli used in single color experiments which could be colorimetrically specified, and experiments with unspecifiable color pictures. Secondly, an emotional response measure employing the three dimensions of pleasure, arousal and dominance, was used to assess the effects of the display dimensions of hue, value, chroma and motif and the subject variable of sex. In addition, a verbal measure of information rate was used to assess the extent to which the display motifs influenced subjects' non-affective (i.e., cognitive responses), and subjects' ability to recognize the display motifs was assessed as well. Thirdly, the problem of stimulus specification was approached through the application of a three-step procedure involving increasing stimulus specificity. These approaches dealt with the specification in terms of (1) the individual color elements making up a display, (2) the quantity of these individual color components, and (3) the distribution or location of these elements across the display surface. The latter specification scheme, which was termed "distribution specification", made use of 24 procedures—some based on accepted artistic views and others of a more abstract nature—for calculating the relationship between the color elements in the displays. The measures which these procedures resulted in were subseqently assessed against subjects' responses on the dimensions of pleasure, arousal, dominance and information rate. Initially, a pilot study with 20 subjects and 16 of the 80 displays was conducted to test the general performance of the response measures and to test whether the displays could be presented in the form of projected slides. The results of this study showed that the general experimental procedure was acceptable but that the projection technique distorted the colors of the displays excessively. Based on the conclusions of the pilot study, a larger study using 82 subjects and the displays as originally constructed was conducted. The results were surprising to the extent that complex color stimuli did not differevery substantially from those elicited by single color stimuli: the color dimension of value influenced the emotional responses to the greatest extent, chroma to a somewhat lesser extent, and hue very little. The motif of the displays, on the other hand, was found to make a substantial difference to the way subjects felt about a display, and the way they assessed it in terms of information rate. Also, it was found that the verbal measure of information rate was a good predictor of how well subjects would recognize a motif. The results of the analysis of stimulus specification in terms of the 24 distribution measures was particularly interesting and gratifying in that several of the measures emerged as strong predictors of responses to the emotional measures and information rate. In particular, the artistically common-sense notions of top-bottom and left-right pictorial balance were prominent, as was the specially constructed measure of contrasts within small sampling areas of the displays. It was concluded, first, that the study had reinforced the findings of many past studies dealing with color and affect, and that it had thrown some new light on some of the controversial and contradictory findings of the past. Secondly, the study had moved the investigation of emotional responses to color pictures and works of art a substantial step closer to realization. Finally, the study had suggested new and promising avenues to follow in the further investigation of colorimetric specification of complex color stimuli. / Graduate and Postdoctoral Studies / Interdisciplinary - Psychology, Education, Architecture / Graduate

A study was done to investigate the well documented heterogeneous nature of the alcoholic population. Loevinger's Ego Development Sentence Completion Scale and Rotter's Locus of Control Questionnaire were used to investigate specific characteristics of the primary and secondary alcoholic subtypes. No differences between primary and secondary alcoholics were found on either the Ego Development or Locus of Control scales. The reasons for and implications of such findings were discussed. / Education, Faculty of / Educational and Counselling Psychology, and Special Education (ECPS), Department of / Graduate

M.Phil. (Personal and Professional Leadership) / in the workplace often cannot work together effectively towards optimal performance. Employees often experience relationship problems in the workplace. Not many realise that a lack of understanding of personality contribute to these problems. Through this study, the objective was to investigate whether a basic understanding of personality types is a factor that can significantly improve workplace relationships in teams and therefore improve effective teamwork and team performance in companies. The motivation for this study was to make a contribution towards helping teams function more effectively, specifically by improving workplace relationships through applying an understanding of personality types in teams. The empirical research method used in this study was a mixed method approach of quantitative and qualitative research. Quantitative research was done through a survey questionnaire that was completed by a sample of respondents (n=183) from companies in the financial industry. This was supplemented by qualitative research by means of focused group interviews (n=16) with team leaders and managers of the survey participants. While the results were not found to be typical of the broader population, enough evidence were found to suggest that employees in the workplace realise the need for working together better in their teams, and seeing the value that personality profiling can bring towards achieving this reality.

This dissertation explored factors related to heterosexual women's use of the label "PMS" and the question of why so many women who say that they have PMS do not meet prospectively-based researcher criteria. Participants were recruited through Vancouver city newspapers and posters for a study of emotional, behavioural, and health patterns. They were screened for hysterectomies, ovariectomies, pregnancy, and chronic illnesses. The 58 women in this study (mean age=34.5) provided daily prospective data over a mean of 15 weeks for 15 variables representing 4 different types of symptoms: mood, relationship, concentration and work performance, and physical and activity symptom types. At the beginning of the study, they completed the Beck Depression Inventory. During a final interview they completed various questionnaires regarding romantic relationships, stress, history of abuse, and attitudes toward menstruation. During her final interview, each woman was asked whether she had ever had PMS and, if so, what she believed caused it. Only 16% of the women met researcher-designated criteria for PMS and 9% met researcher-criteria for PDD (Premenstrual Dysphoric Disorder, APA, 1994), but 60% believed that they currently had PMS. There was very little match between researcher-designations and self-designations for any of the 15 prospective variables. Instead, consistent with schematic theories of PMS, it was women's attitudes toward menstruation that were most strongly related to self-designation. Women who said that they currently had PMS were more likely to view menstruation as debilitating, natural, and predictable. They had higher depression scores and reported more frequent and enduring experiences of anger. More of these women reported having been emotionally abused as an adult, emotionally abused as a child, and physically abused by a past or current romantic partner. Although women who said that they had never had PMS perceived menstruation as more bothersome, they believed that women should be able to ignore it. Current-Say and Never-Say women did not significantly differ for other forms of abuse, partner satisfaction, or daily uplifts. Daily hassles did not reach statistical significance at the multivariate level, but univariate significance indicated that Never-Say women might experience fewer daily hassles than Current-Say women. The prospective data were analyzed idiographically using prediction analyses. Current-Say women demonstrated more uncharacteristic cyclicity during the midcycle phase across the 15 variables and 4 symptom types. Never-Say women showed more uncharacteristic cyclicity during the postmenstrual phase. No differences were found for other phases. These findings, and other results, argue against the use of simple premenstrual-postmenstrual and premenstrual-intermenstrual difference measures in diagnosing PMS or PDD. It was concluded that, although the women's self-designations did not match their prospective data, they could not be explained simply as a mislabelling of negative cyclicity in other phases. There was also mixed evidence for the hypothesis that women's reports of having "PMS" were part of a generalized over-reporting of symptomatology or dissatisfaction. Possible explanations for women's self-designations are discussed, including schematic representations of both menstruation and general illness and a "contrast hypothesis" modified from the version proposed by McFarlane and Williams (1994). This contrast hypothesis suggests that cyclical changes occurring during a particular non-premenstrual phase are related to women's self-designations according to (a) the timing of these changes relative to the visible and salient menses, in conjunction with (b) their attitudes toward menstruation. Close to half (45%) the women who said that they had experienced PMS attributed their perceived PMS to a bidirectional model of physiology and environment, and 58% of the women saying that they had PMS believed that it happened occasionally rather than inevitably. Overall, the women's representations of menstrual cyclicity were neither simple reflections of cultural stereotypes nor pervasively negative, but rather diverse and complex. The results that link depression, anger, and specific forms of abuse to self-designations suggest that women saying that they have PMS are indicating difficulties that may or may not be menstrually-related. Researchers and other professionals need to assess the nature of those difficulties in women presenting with "PMS". / Arts, Faculty of / Psychology, Department of / Graduate

The present study is an attempt to determine the effect of anxiety, hostility, and depression on responses to the Rotter Incomplete Sentences Blank when it is scored according to the manual instructions. Whether the score fluctuates or not will have implications on how psychologists should use this test as a diagnostic tool.

Procedures based on item response theory (IRT) are widely accepted for solving various measurement problems which cannot be solved using classical test theory (CTT) procedures. The desirable features of dichotomous IRT models over CTT are well known and have been documented by Hambleton, Swaminathan, and Rogers (1991). However, dichotomous IRT models are inappropriate for situations where items need to be scored in more than two categories. For example, in performance assessments, most of the scoring rubrics for performance assessment require scoring of examinee's responses in ordered categories. In addition, polytomous IRT models are useful for assessing an examinee's partial knowledge or levels of mastery. However, the successful application of polytomous IRT models to practical situations depends on the availability of reasonable and well-behaved estimates of the parameters of the models. Therefore, in this study, the behavior of estimators of parameters in polytomous IRT models were examined. In the first study, factors that affected the accuracy, variance, and bias of the marginal maximum likelihood (MML) estimators in the generalized partial credit model (GPCM) were investigated. Overall, the results of the study showed that the MML estimators of the parameters of the GPCM, as obtained through the computer program, PARSCALE, performed well under various conditions. However, there was considerable bias in the estimates of the category parameters under all conditions investigated. The average bias did not decrease when sample size and test length increased. The bias contributed to large RMSE in the estimation of category parameters. Further studies need to be conducted to study the effect of bias in the estimates of parameters on the estimation of ability, the development of item banks, and on adaptive testing based on polytomous IRT models. In the second study, the effectiveness of Bayesian procedures for estimating parameters in the GPCM was examined. The results showed that Bayes procedures provided more accurate estimates of parameters with small data sets. Priors on the slope parameters, while having only a modest effect on the accuracy of estimation of slope parameters, had a very positive effect on the accuracy of estimation of the step difficulty parameters.

There is considerable evidence to show that computerized-adaptive testing (CAT) and multi-stage testing (MST) are viable frameworks for testing. With many testing organizations looking to move towards CAT or MST, it is important to know what framework is superior in different situations and at what cost in terms of measurement. What was needed is a comparison of the different testing procedures under various realistic testing conditions. This dissertation addressed the important problem of the increase or decrease in accuracy of ability estimation in using MST rather than CAT. The purpose of this study was to compare the accuracy of ability estimates produced by MST and CAT while keeping some variables fixed and varying others. A simulation study was conducted to investigate the effects of several factors on the accuracy of ability estimation using different CAT and MST designs. The factors that were manipulated are the number of stages, the number of subtests per stage, and the number of items per subtest. Kept constant were test length, distribution of subtest information, method of determining cut-points on subtests, amount of overlap between subtests, and method of scoring total test. The primary question of interest was, given a fixed test length, how many stages and many subtests per stage should there be to maximize measurement precision? Furthermore, how many items should there be in each subtest? Should there be more in the routing test or should there be more in the higher stage tests? Results showed that, in general, increasing the number of stages from two to three decreased the amount of errors in ability estimation. Increasing the number of subtests from three to five increased the accuracy of ability estimates as well as the efficiency of the MST designs relative to the P&P and CAT designs at most ability levels (–.75 to 2.25). Finally, at most ability levels (–.75 to 2.25), varying the number of items per stage had little effect on either the resulting accuracy of ability estimates or the relative efficiency of the MST designs to the P&P and CAT designs.

Tests consisting of both multiple-choice and constructed-response items have gained in popularity in recent years. The evidence shows that many assessment programs have administered these two item formats in the same test. However, linking these two item formats on a common scale has not been thoroughly studied. Even though several methods for linking scales under item response theory (IRT) have been developed, many studies have addressed multiple-choice items only and only a few studies have addressed constructed-response items. No linking studies have addressed both item formats in the same assessment. The purpose of this study was to investigate the effects of several factors on the accuracy of linking item parameter estimates onto a common scale using the combination of the three-parameter logistic (3-PL) model for multiple-choice items with the graded response model (GRM) for constructed-response items. Working with an anchor-test design, the factors considered were: (1) test length, (2) proportion of items of each format in the test, (3) anchor test length, (4) sample size, (5) ability distributions, and (6) method of equating. The data for dichotomous and polytomous responses for unique and anchor items were simulated to vary as a function of these factors. The main findings were as follows: the constructed-response items had a large influence in parameter estimation for both types of item formats. Generally, the slope parameters were estimated with small bias but large variance. Threshold parameters were also estimated with small bias but large variance for constructed-response items. However, the opposite results were obtained for multiple-choice items. For the guessing parameter estimates, the recovery was relatively good. The coefficients of transformation were also relatively well estimated. Overall, it was found that the following conditions led to more effective results: (1) a long test, (2) a large proportion of multiple-choice items in the test, (3) a long anchor test, (4) a large sample size, (5) no ability differences between the groups used in linking the two tests, and (6) the method of concurrent calibration. At the same time, more research will be necessary to expand the conditions, such as the introduction of multidimensional data, under which linking of item formats to a common scale is evaluated.

Computer-based testing is becoming popular with credentialing agencies because new test designs are possible and the evidence is clear that these new designs can increase the reliability and validity of candidate scores and pass/fail decisions. Research on MST to date suggests that the measurement quality of MST results is comparable to full-fledged computer-adaptive tests and improved over computerized fixed-form tests. MST's promise dwells in this potential for improved measurement with greater control than other adaptive approaches for constructing test forms. Recommending use of the MST design and advising how best to set up the design, however, are two different things. The purpose of the current simulation study was to advance an established line of research on MST methodology by enhancing understanding of how several important design variables affect outcomes for high-stakes credentialing. Modeling of the item bank, the candidate population, and the statistical characteristics of test items reflect an operational credentialing exam's conditions. Studied variables were module arrangement (4 designs), amount of overall test information (4 levels), distribution of information over stages (2 variations), strategies for between-stage routing (4 levels), and pass rates (3 levels), for 384 conditions total. Results showed that high levels of decision accuracy (DA) and decision consistency (DC) were consistently observed, even when test information was reduced by as much as 25%. No differences due to the choice of module arrangement were found. With high overall test information, results were optimal when test information was divided equally among stages; with reduced test information gathering more test information at Stage 1 provided the best results. Generalizing simulation study findings is always problematic. In practice, psychometric models never completely explain candidate performance, and with MST, there is always the potential psychological impact on candidates if test difficulty shifts are noticed. At the same time, two findings seem to stand out in this research: (1) with limited amounts of overall test information, it may be best to capitalize on available information with accurate branching decisions early, and (2) there may be little statistical advantage in exceeding test information much above 10 as gains in reliability and validity appear minimal.