A note on preliminary tests of equality of variances

Preliminary tests of equality of variances used before a test of location are no longer widely recommended by statisticians, although they persist in some textbooks and software packages. The present study extends the findings of previous studies and provides further reasons for discontinuing the use of preliminary tests. The study found Type I error rates of a two‐stage procedure, consisting of a preliminary Levene test on samples of different sizes with unequal variances, followed by either a Student pooled‐variances t test or a Welch separate‐variances t test. Simulations disclosed that the twostage procedure fails to protect the significance level and usually makes the situation worse. Earlier studies have shown that preliminary tests often adversely affect the size of the test, and also that the Welch test is superior to the t test when variances are unequal. The present simulations reveal that changes in Type I error rates are greater when sample sizes are smaller, when the difference in variances is slight rather than extreme, and when the significance level is more stringent. Furthermore, the validity of the Welch test deteriorates if it is used only on those occasions where a preliminary test indicates it is needed. Optimum protection is assured by using a separate‐variances test unconditionally whenever sample sizes are unequal.     

Graduate pre-selection: A research note

Information about educational, work and leisure achievements discriminated between graduate application forms which were accepted for further consideration and those which were rejected, as did personal background features and style of completion of the application form. Variation between types of organization, individual organizations and selectors from the same organization was observed.     

Parity and disparity among three measures of fear of crime: A research note

Research on fear of crime often concerns problems associated with measuring fear of crime. Most of the literature in this area critiques “global measures,” which are the most commonly used, and suggests movement to more specific measures of fear of crime. Although much has been said about the use of alternative measures of fear, seldom has research provided comparative analyses of the alternative measures. Using a conceptual model based on previous research, the present study illustrates the differences among findings using three measures of fear of crime: global fear, fear of property crime, and fear of violent crime. This is achieved by employing each of the three measures of fear as the dependent variable in identical models. Comparing the results, the parity and disparity of these measures are demonstrated. The results suggest that conflicting findings from previous research may be a consequence of the types of measures used in the analyses. Thus, it is theoretically important to understand what dimensions of fear are being measured by our operational procedures.     

The lexical decision list task: A research note

The lexical decision list task, a paper-and-pencil version of the standard lexical decision task, is introduced and described. Data are reported demonstrating that subjects' responses to manipulations of phonological confusability, stimulus difficulty, and case presentation are analogous to those obtained with the use of the standard task.     

Stimulus intensity control and personality: A research note

Previous research has suggested that variations in augmenting-reducing, extraversion, and sensation seeking might be explained in terms of a common stimulus intensity control mechanism. However, empirical tests of this hypothesized relationship have produced equivocal findings. The present study attempts to clarify the issue by analyzing scores for extraversion, sensation seeking, and both cortical and questionnaire measures of augmenting-reducing obtained from the same sample of subjects. Apart from a significant positive correlation between the questionnaire index of reducing and sensation seeking, which have not previously been investigated, the results show no evidence of a systematic relationship among the variables.     

Southern traditionalism and sex-role ideology: A research note

Sociological literature, as well as popular folklore, suggests the existence of a Southern regional subculture, of which sex-role traditionalism is an integral part. To test the empirical validity of this assumption, the present study employed the 1975 National Fertility Study data, a probability sample of 3,403 currently married white women, of whom eight sex-role questions were asked. A factor analysis of these questions was used to calculate a sex-role summary score for each respondent. Wife's employment in 1975 and her attainment of high levels of education were the two strongest predictors of a nontraditional sex-role score. The sex-role attitudes of Southern and non-Southern women did not differ significantly. The implications for research and policy are discussed.An earlier version of this article was presented at the Annual Meeting of the Southern Sociological Society, Atlanta, Georgia, April 4–7, 1979. This project was supported by the Center for the Study of Population and Natural Resources at Michigan State University and by Michigan Agricultural Experiment Station Project No. 3243 S. This report is Michigan Agricultural Experiment Station Journal Article Number 11107. The authors thank Merwyn R. Nelson and an anonymous reviewer for constructive comments on an earlier draft.     

School rebellion: a research note

Pathanalysis is applied to survey data bearing on Arthur Stinchcombe's model of student rebellion. The linear causal model hypothesizes that school rebellion is due directly to expressive alienation and indirectly to strain in three systems: poor articulation regarding status allocation, claims to cultural symbols of adult rights, and unrealizable internal standards. Support for the model is not obtained. An alternate model is tested and verified in which political efficacy is an important antecedent of school disruption. The major implication is that school rebellion is not a response to the school experience itself. Other findings are presented concerning the scalability of school disruption items, the nature of expressive alienation and its components, and the impact of sex, race, and neighborhood on school rebellion.     

Gender and gender-role categorization of emergent leaders: A critical review and comprehensive analysis

The influences of gender and gender role on emergent leadership in initially leaderless groups were examined. Subjects were multi-ethnic, multi-racial domestic and international M.B.A. students engaged in gender-neutral group projects throughout the term. The BSRI was administered near the beginning of the term and leader emergence measures were collected at the end of the term after many weeks of group interaction. Using leader emergence measures which allowed for the possibility of multiple group leaders, both masculine and androgynous group members emerged as leaders. In groups with one sole leader, masculine types were more likely to be nominated as the single leader than any other gender-role type. There were no gender effects on any measure of leader emergence.     

The occupation of funeral director: A research note on work orientations

Factor analysis of responses of funeral directors to work orientation items is interpreted first as specifying major orientational themes and second as to how well the theoretical value components of professionalism and a business orientation hold together empirically. The analysis revealed that the professionalism model as operationalized in this study has no single underlying dimension which would define this orientation. It is thus suggested that the meaning of professionalism be altered to refer only to involvement in the occupational structure and not include service and autonomy. The business orientation model, however, appears to be fairly well supported by the data.     

A research note on the new SVIB Holland scales

The Strong Vocational Interest Blanks (SVIB) of 93 management students were scored using six new scales which correspond to Holland's six occupational types. The scores from both the 14-item and the 20-item scales were correlated with actual scores on the Vocational Preference Inventory. The 14-item and the 20-item scales were also correlated with each other. In addition, the SVIB-derived Holland scales were scored using a simplified procedure, and the results were compared with the scoring procedure used by Campbell.     

Religiousness and happiness in three nations: a research note

Several empirical investigations have found positive correlations between religiousness and happiness and this is typically seen as a proof of a universal beneficial effect of religion. Yet most studies were done in the USA and the balance of effects may be different in other countries, in particular in countries where the social and economic functions of churches are less prominent and where believers are in the minority.This hypothesis was checked in an analysis of the World Values Survey in three countries: the USA, the Netherlands and Denmark in 2000. Seven self report indicators of religiousness were used. Happiness was measured using a single question on life-satisfaction.The correlations between religiousness and happiness appear to be positive but weak. The correlations are stronger in the USA (average + 0.13, most significant) than in The Netherlands (average + 0.05, none significant) and Denmark (average + 0.05, none significant).The pattern of correlation is not much different among people who might benefit more from the support of religion, such as widows. Apparently religiousness is not always a blessing.     

Relationship of testosterone to men's family functioning at mid-life: A research note

Sociobiological theory suggests that aggression coupled with altruism is highly adaptive for males in all social systems, including the family, whereas sex-role convergence theory indicates that lower levels of aggression are most adaptive for men, particularly at mid-life. The purpose of this study was to determine the role that testosterone, as a proxy for aggression, plays as an adaptive mechanism in the lives of middle-aged males. The sample used in this study consisted of 37 middle-aged (39–50 years of age) males employed in professional occupations in a midwestern metropolitan area. Testosterone levels were regressed on factors typically found to be related to male satisfaction with family life at mid-life: marital satisfaction, parent-adolescent communication, amount of emotional expressiveness, and androgynous characteristics; recent positive and negative stressors and trait anxiety were statistically controlled. Results of step-wise multiple regression analysis indicated that low levels of testosterone were significantly (R = .34, P < .01) related to enhanced marital and parental relationships and androgynous behaviors; however, emotional expressiveness was significantly related to high levels of testosterone.     

Variations in patterns of lateral asymmetry among dextrals

Evidence supporting individual variations in patterns of hemispheric involvement in the recognition of visuo-spatial and verbal stimuli among dextrals is reported. In Experiment 1, subjects' asymmetry scores on a task that was nonlateralized for the group as a whole were significantly correlated with their asymmetry scores on right-hemisphere-specialized tasks, including face recognition. In Experiment 2, subjects' asymmetry scores on a task that was nonlateralized for the group as a whole were significantly correlated with their asymmetry scores on a left-hemisphere-specialized word recognition task. These results suggest that individual dextrals' asymmetry scores on lateralized tasks are a joint function of a subject's underlying hemispheric specialization for that task and stable individual variations in asymmetric hemispheric reliance.