The Impact of Mode of Administration on the Equivalence of a Test Battery: A Quasi‐Experimental Design

Two quasi‐experimental studies were conducted to examine the equivalence of an online unsupervised test battery (ICES ability, interests and personality tests) in relation to supervised paper administration and to test whether differences were because of the lack of supervision, the test being online as opposed to on a PC, or the computerization of the test. In the first study, data from a sample of 86 employed individuals suggested a lack of equivalence on ability scales between online unsupervised and paper supervised testing modes. Effect sizes ranged from .04 to .49 and corrected coefficients of equivalence and stability from .65 to .81. Study two used a more controlled design as well as obtaining data on personality and interest scales. Data from 118 college and university students were collected, with 64 students replicating the design in study 1 and 54 tested in supervised paper and supervised PC modes. For the ability scales, a similar pattern to study 1 was seen between supervised paper and unsupervised online testing as well as between supervised paper and supervised PC testing. Data for interest and personality scales were more positive showing lower effect sizes and higher coefficient values across testing modes. When ability test data from the online testing group were compared with data from the PC group, results showed that the scales were affected by the computerization process rather than the lack of supervision or because they were in online as opposed to PC mode.     

Racial differences in socially desirable responding in selection contexts: magnitude and consequences

Two studies were conducted to examine the magnitude and consequences of racial differences on social desirability (SD) scales. Study 1 included 1,063 job applicants, and Study 2 included 3 sets of incumbents (total N = 534). In both studies, participants were administered several personality measures and an SD scale. Across all samples, Whites scored lower on the SD scale than Blacks (average d = .37), Hispanics (average d = .47), and Asians (average d = 1.04), but these differences were not observed on the personality scales. The consequence of differences in socially desirable responding (SDR) is that fewer minority group members would be selected if SD scales were used to derive cut scores to eliminate individuals from the applicant pool or if the scales were used to correct personality test scores for SDR. However, applying the SD correction did not affect the validity of the personality test for any of the racial groups. Overall, our findings suggest that researchers and practitioners should consider the use of SD scales very carefully, as their use may have unintended consequences. These studies also demonstrate a need to closely examine the construct validity of SD measures across diverse groups.     

The supervisory alliance: Comparison of measures and implications for a supervision toolkit

Background Clinical supervision lies at the core of the professional development of clinicians providing services within a range of service settings. The supervisory alliance is regarded as the most important construct in effective supervision which underlies the importance of metrics to assess this construct. Aims materials and Method A cross‐sectional self‐report questionnaire design was used to examine clinical supervision constructs. The study used two scales measuring the supervisory alliance (SRQ and SWAI‐T), a scale measuring supervision satisfaction (SSQ) and a scale measuring disclosure (TDS). The questionnaires were administered to 127 clinical psychology trainees who were undertaking a practicum placement. Results Both scales measuring the alliance were positively correlated with satisfaction with supervision and disclosure, with similar sizes of correlation for each alliance scale. There was also a large, positive correlation between the two scales of measuring the alliance. Discussion Both scales were designed for use in supervision and are related to supervision outcome measures in the expected directions. Conclusion The SRQ has some distinct advantages over the SWAI‐T that lend the SRQ to be potentially more useful in clinical and research settings.     

A content and correlational analysis of seven locus of control scales

Seven locus of control scales—two designed for adults and five for children—were administered to about 200 adolescents. A content analysis revealed very little overlap between the scales which had practically no effect on the correlations between them. Correlations between the five children’s locus of control scale scores were highly significant, and nearly all greater than .50, but the two adult scales showed very little significant correlations either with each other, or any of the children’s scales. A number of demographic variables were correlated with total scale scores of which age was the most significant. The results are discussed in terms of the multidimensional nature of the concept, psychometric evaluation of locus of control beliefs in different age groups and the specificity of the concept in general.     

A CRITICAL ANALYSIS OF STUDIES COMPARING BEHAVIORALLY ANCHORED RATING SCALES (BARS) AND OTHER RATING FORMATS1

All published studies comparing behaviorally anchored rating scales (BARS) with other rating formats were reviewed. Results indicate that BARS and other formats appear to differ relatively little (if at all) with regard to psychometric characteristics (leniency, halo, interrater agreement, ratee discriminability, validity and “accuracy”). Other variables, such as degree of rater participation in scale construction and rater preferences for different rating scale formats, also exhibited relatively little differences between BARS and other formats. A close examination of the studies, however, reveals numerous methodological problems confounding interpretations of results, making conclusions about the superiority or inferiority of BARS quite tenuous. Suggestions are made for more appropriate research directions and questions which seem more appropriate for rating scale investigations.     

Sex of Supervisor, Sex of Subordinate, and Subordinate Job Satisfaction

It was hypothesized in the present research that there would be significant differences in the job satisfaction of subordinates depending upon the interaction between their sex and the sex of their supervisors. Using job satisfaction data from 139 employees, it was found that: (a) female subordinates supervised by male supervisors were less satisfied with supervision than those supervised by women, (b) females working under males reported a lower level of satisfaction with their co-workers than females supervised by females or males supervised by males, and (c) women who had female supervisors indicated a higher level of satisfaction with their work than men with male supervisors. Contrary to traditional stereotypes of women supervisors, the results suggested that these stereotypes might not be important in long-term, real-world situations.     

A Survey of Clinical Supervision in Genetic Counseling

Three hundred and thirty-five full members of the National Society of Genetic Counselors completed a survey concerning the nature of clinical supervision for genetic counseling students. Slightly over half (54.3%, n = 182) had provided clinical supervision within the past 5 years. Of those who supervised, 66.7% had 5 years or less supervision experience, and 55% had supervised 10 or fewer students. The majority became supervisors because they enjoy teaching and contributing to the profession. Common reasons for not supervising include no local graduate program, and had never been asked to supervise. Live supervision is the most prevalent student evaluation method: every supervisor reported providing one-on-one oral feedback, and 47.3% indicated that they always provide feedback immediately following a counseling session. The most frequent challenges involve students who lack technical knowledge and who fail to incorporate feedback. Training, policy, and research implications are discussed.     

The development of the Meaning in Life Index (MILI) and its relationship with personality and religious behaviours and beliefs among UK undergraduate students

The scales available for assessing meaning in life appear to be confounded with several related constructs, including purpose in life, satisfaction with life, and goal-directed behaviour. The Meaning in Life Index (MILI), a new instrument devised as a specific measure of meaning in life, was developed from responses to a pool of 22 items rated by a sample of 501 undergraduate students in Wales. The nine-item scale demonstrated sufficient face validity, internal consistency, and scale reliability to commend the instrument for future use. With respect to personality, the MILI scores were most strongly predicted by neuroticism (negatively), and less strongly by extraversion (positively) and psychoticism (negatively). With respect to several religious behavioural variables, those who attended church at least weekly returned significantly higher MILI scores than those who attended church less frequently. Intrinsic religiosity was the only orientation to be significantly associated with the MILI scale scores, although the magnitude of the association was smaller than anticipated. These results suggest that meaning in life is associated more strongly with individual differences in personality than with specific religious behaviours and attitudes. The implications of these results are discussed in terms of individual's personal values and attitudes that might underlie their experience of a meaning in life.     

Sun and water: on a modulus-based measurement of happiness

Most happiness researchers use semantic differential or Likert scales to assess happiness. Such conventionally used scales are susceptible to scale renorming (interpretation of scales differently in different contexts) and can produce a specious relativism effect (e.g., rating a low-income person happier than a high-income person in situations where the low-income person is not happier). Building on related psychophysical measurements, the authors propose a simple, survey-friendly, modulus-based scale of happiness and show that it is less susceptible to specious relativism than conventional rating scales but can still catch genuine relativism (e.g., rating a low-income person to be happier than a high-income person in situations where the low-income person is indeed happier).     

Measurement of depression: Comparison between self-reports and clinical assessments of depressed outpatients

Agreement among several depression scales was investigated as regard the relative influences of administration mode (self-rating or clinical rating) and scale content. The Beck Depression Inventory (BDI), the Self-Rating Depression Scale (SDS), the Hamilton Rating Scale for Depression (HRSD), and three corresponding scales with identical structure and content but the alternative administration mode were administered to 47 outpatients with diagnoses of DSM-III major depression disorders. Correlations between the total scores and the degrees of association between corresponding items of different scales were calculated. The results suggest that differences in content contribute more to inter-scale discrepancy than differences in administration mode. The implications for the evaluation of outpatients with major depression are discussed.     

Confirmatory factor analyses of the full and short versions of the Marlowe-Crowne Social Desirability Scale

Over the years, researchers have developed various short versions of the Marlowe-Crowne Social Desirability Scale (D. P. Crowne & D. Marlowe, 1960). The authors used confirmatory factor analyses (J. L. Arbuckle, 1997) as well as item and scale analyses to evaluate the adequacy of the full version and various short versions. Overall, the results from 232 Canadian undergraduates showed (a) that all the short versions in the present study are a significant improvement in fit over the 33-item full scale and (b) that W. M. Reynolds's (1982) Forms A and B are the best fitting short versions. No gender differences were found for the full scale or any of the short versions. The results show that the full scale could be improved psychometrically and that the psychometrically sound short versions should be available because they require less administration time than the full scale.     

Effects of Supervisor Style and Empathy on Counselor Perceptions in Simulated Counseling

Effects of supervisor style (techniques type and counseling type) and supervisor empathy upon counselor perceptions in a supervision analogue were examined. Three groups of 14 students heard two sets of recorded client statements and responded as counselors. The two experimental (E) groups had 20-minute conferences with supervisors between the first and second sets. Controls (C) received no supervision. Counselor perceptions were measured by questionnaire responses. Although subjects supervised by an emphasis on techniques rated themselves less relaxed and less confident about the meaning of empathy than those with an emphasis on counseling, their performance on the criterion of empathy was higher. Both E groups perceived themselves as higher than C's in amount learned, but they were no higher on the performance measure. No differences were found for reactions to supervision (items given E's only) or reactions to participation in the experiment. Empathy of supervisors did not appreciably affect counselor perceptions.     

Internal external control of reinforcement as a determinant of study habits and academic attitudes

This study tested Rotter's hypothesis that internals would show more achievement-striving behaviors than externals using a direct measure of such behavior. Rotter's I-E scale and the Survey of Study Habits and Attitudes (SSHA) were administered to 123 male and 130 female undergraduate subjects. Significant sex differences were found on the SSHA scales but not on the I-E scale. Results indicated significant differences between internals and externals within each sex in terms of the average SSHA profiles and the individual SSHA scales. Using H. L. Mirels' (Journal of Consulting and Clinical Psychology, 1970, 34, 226–228) I-E subscales, similar results were obtained with the personal-control scale but not the political-control scale. Results provide additional evidence supporting the multidimensional interpretation of the I-E scale.     

Confirmatory Factor Analyses of the Full and Short Versions of the Marlowe-Crowne Social Desirability Scale

Abstract

Over the years, researchers have developed various short versions of the Marlowe-Crowne Social Desirability Scale (D. P. Crowne & D. Marlowe, 1960). The authors used confirmatory factor analyses (J. L. Arbuckle, 1997) as well as item and scale analyses to evaluate the adequacy of the full version and various short versions. Overall, the results from 232 Canadian undergraduates showed (a) that all the short versions in the present study are a significant improvement in fit over the 33-item full scale and (b) that W. M. Reynolds's (1982) Forms A and B are the best fitting short versions. No gender differences were found for the full scale or any of the short versions. The results show that the full scale could be improved psychometrically and that the psychometrically sound short versions should be available because they require less administration time than the full scale.     

The Role of Parental Attitudes and Monitoring in the Risk Exposure of Young Children1

The parental monitoring literature debates the role of parental activity, as opposed to parental knowledge in childhood risk taking. We examined parents’ attitudes and behaviors regarding roadside supervision of children aged 4 to 6 years using self‐report and observational measures. Although parents perceived sons as less careful than daughters, they reported no sex differences in the age children would be allowed out unsupervised. There were no self‐report differences in supervision according to socioeconomic status. However, roadside observations showed that boys and children from a lower socioeconomic area received less supervision than girls or those from a higher socioeconomic area. Such findings contribute to our understanding of the sex and socioeconomic differences in accident involvement and the role of parental monitoring.     

Psychological and neurological predictors of abusive supervision

Although the negative effects of abusive supervision are well documented, less is known about the individual differences that drive supervisors to be abusive. We use a self‐control perspective to understand the unique roles of both psychological and neurological characteristics of supervisors in the prediction of abusive behavior. Specifically, we find a positive relationship between narcissism and abusive supervision, whereas political skill and intrinsic neurological connectivity in executive control regions of the brain negatively relate to abusive supervision. Our results further show that the relationship between narcissism and abusive supervision diminishes for those who are strong in terms of political skill. In addition, neurological executive control moderates the relationship between political skill and abusive behavior. When connectivity in executive control regions of the brain is lower, political skill becomes a more important factor in reducing the display of abusive behavior. Overall, we demonstrate how combinations of characteristics (both psychological and neurological) can provide a more in‐depth understanding of the emergence of abusive supervision.