A comparison of two procedures for calculating discriminant function coefficients

Alternative procedures for calculating discriminant function coefficients have been illustrated in reported research. One method proceeds from data which has been expressed in deviation score units, whereas the other method implies that the data has been expressed in standard units. It is shown algebraically that both procedures yield identical discriminant function coefficients, and therefore, the method involving less mathematical manipulation and computation is preferable.     

The team‐level model of climate for innovation: A two‐level confirmatory factor analysis

The level structure of West's (1990) four‐factor model of team climate for innovation was assessed by means of multi‐level confirmatory factor analysis (MCFA). The sample consisted of 1,487 individuals (195 teams) from a wide range of professions. Results showed that a considerable portion of the variance in the data was explained on the team level with intra‐class correlations ranging from .30 to .39. Furthermore, the results demonstrated that the overall measurement model fitted the data well at both the team and individual levels, while the factor loadings were slightly different across the levels with item loadings showing partial invariance. Results from confirmatory factor analyses conducted on separate levels, however, showed that the four‐factor model displayed the best fit to the data for both individual and team levels. A second‐order one‐factor model also fitted the data well on both levels. The results indicate that the team climate for innovation model can be used as a team‐level consensus model of team climate for innovation.     

A comparison of two college classroom testing procedures: required remediation versus no remediation

Forty one subjects from a 10-week introductory course in Educational Psychology were randomly divided into two experimental groups. All students took weekly quizzes over content material. Members of one group received little or no academic credit if they performed at less than 90% on a weekly quiz, but could earn additional credit by taking a weekly remedial quiz. Members of the second group also took the initial weekly quizzes, but retained their raw scores and were not permitted to take the weekly remedial quizzes. Performance on a 100-item multiple-choice comprehensive final revealed a statistically significant and educationally important difference between the two groups, the required-remediation group scoring an average of one-half letter grade higher.     

A comparison of several procedures for eliminating behavior

The pecking response of pigeons was reinforced when a minimum period of time had elapsed since the last response (DRL schedule of food-reinforcement). Punishment, satiation, extinction, and stimulus change were employed separately to reduce responding. When the effects of the four procedures were compared, punishment was found capable of producing a more immediate, complete and long lasting response reduction than the others. Punishment had its maximum effect on the responses that were least relevant to reinforcement. The punishment reduced the frequency of the short inter-response times to a greater extent than did either extinction or satiation. In this way, punishment actually increased the efficiency of the DRL responding.     

A comparison of procedures for programming noncontingent reinforcement schedules

We compared two methods for programming and thinning noncontingent reinforcement (NCR) schedules during the treatment of self-injurious behavior (SIB). The participants were 3 individuals who had been diagnosed with mental retardation. Results of functional analyses indicated that all participants' SIB was maintained by positive reinforcement (i.e., access to attention or food). Following baseline, the effects of two NCR schedule-thinning procedures were compared in multielement designs. One schedule (fixed increment) was initially set at fixed-time 10-s reinforcer deliveries and was also thinned according to fixed-time intervals. The other schedule (adjusting IRT) was initially determined by participants' baseline interresponse times (IRTs) for SIB and was thinned based on IRTs observed during subsequent treatment sessions. Results indicated that both schedules were effective in initially reducing SIB and in maintaining response suppression as the schedules were thinned.     

Comparison of closed testing procedures for pairwise testing of means

A Monte Carlo simulation was conducted to compare 9 pairwise multiple comparison procedures. Procedures were evaluated on the basis of any-pair power and all-pairs power. No procedure was found to be uniformly most powerful. A modification due to A. J. Hayter (1986) of Fisher's least significant difference was found to provide the best combination of ease of use and moderately high any-pair power in most cases. Pilot or exploratory studies can expect good power results with this relatively simple procedure. The greatest all-pairs power was usually provided by 1 of 2 partition-based versions of E. Peritz's (1970) procedure. Confirmatory studies will require such complex methods but may also need larger sample sizes than have been customary in psychological research.     

Cognitive testing in non‐demented Turkish immigrants ‐ comparison of the RUDAS and the MMSE

Nielsen, T. R. Vogel, A., Gade, A. & Waldemar, G. (2012). Cognitive testing in healthy Turkish immigrants ‐ comparison of the RUDAS and the MMSE. Scandinavian Journal of Psychology 53, 455–460. Methods for culturally and linguistically appropriate cognitive testing of elderly minority populations are lacking in Europe. The aim of this study was to compare performance on the Rowland Universal Dementia Assessment Scale (RUDAS) and the Mini Mental State Examination (MMSE) in Turkish immigrants in Denmark and determine the impact of demographic and health‐related variables on test performance. A sample of non‐demented community‐dwelling Turkish immigrants was recruited from the greater Copenhagen area. All participants completed a structured interview regarding demographic, physical and mental health status, as well as measures of depression and acculturation, and cognitive testing with the RUDAS and the MMSE. A total of 76 non‐demented participants aged 50 or more were included in the study. The mean performance on the RUDAS and the MMSE was 26.8 (SD 2.4) and 23.7 (SD 4.3), respectively. In group comparisons, correlation analyses and regression analyses, level of schooling represented a more significant variable for RUDAS and MMSE performance than any other variable. However, the impact of schooling was considerably more pronounced on the MMSE and the test was not found to be a valid measure of general cognitive function in subjects with less than five years of schooling. Although not entirely free of educational bias, the RUDAS can be a valuable supplement to the MMSE for assessment of general cognitive function in Turkish minority populations.     

Construct and Criterion‐Related Validation of a Measure of Emotional Reasoning Skills: A two‐study investigation

In a two‐study investigation, the construct and criterion‐related validities of the Test of Emotional Intelligence, an ability‐based measure of emotional reasoning skills, were examined. In Study 1, as hypothesized, emotional reasoning skills were related to emotion recognition ability (i.e., as measured by Diagnostic Analysis of Nonverbal Accuracy), emotional empathy, and three measures of successful social functioning (i.e., social astuteness, interpersonal influence, and apparent sincerity) assessed by peers, after controlling for personality and occupational environment characteristics. In Study 2, results from a predictive validation study demonstrated that Emotional Intelligence explained additional variance in overall job performance ratings beyond general mental ability and personality traits. Contributions and implications of this research investigation, limitations, and directions for future research are discussed.     

A comparison of self-instructional checking procedures for remediating mathematical deficits

We examined the relative effectiveness of three procedures for teaching long multiplication/division to seven adolescents with learning disabilities: no-checking, end-checking, and multi-checking. During training, each subject was taught by modelling and imitation to verbalize self-instructions in the form of a strategy while solving the problems. The relative effects of the checking and no-checking procedures on accuracy and rate of problems completed were examined in an alternating treatments design. The best treatment was then given alone and a reversal was implemented six weeks later, followed by a return to the best treatment during a final phase. Irrespective of the procedure used, the subjects' accuracy improved while the rate of problems completed decreased. These effects were greatest with the multi-checking procedure for six of the seven subjects. Generalization to untaught problems of various levels of complexity occurred under all procedures. Though maintenance effects were seen during the follow-up, accuracy was generally higher and more reliable when the subject's best checking procedure was reinstated. It is suggested that error detection and correction were important for maintaining high levels of accuracy and that these operated differentially in the three procedures to produce the differing levels of accuracy. The role of other factors such as pre-skill knowledge, complexity of the problem and prior reinforcement history are also considered.     

A comparison of procedures for eliminating self-injurious behavior of retarded adolescents

An attempt was made to eliminate the self-injurious behaviors of four institutionalized, profoundly retarded adolescents. Some of the behaviors studied were: face-slapping, face-banging, hair-pulling, face-scratching, and finger-biting. Three remediative approaches to self-injurious behavior were compared. Elimination of all social consequences of the self-injurious behavior was not effective with the two subjects with whom it was attempted. The same two subjects were exposed to a procedure involving reinforcement of non-self-injurious behavior which was ineffective under no food deprivation and was effective with one of the two subjects under mild food deprivation. Electric-shock punishment eliminated the self-injurious behaviors of all four subjects with whom it was attempted. The results suggested that punishment was more effective than differential reinforcement of non-self-injurious behavior which, in turn, was more effective than extinction through elimination of social consequences. However, the effects of the punishment were usually specific to the setting in which it was administered. In order to eliminate the self-injurious behaviors of severely retarded children, it is apparently necessary to carry out the treatment in many of the settings in which it occurs.     

Sensitivity to initial values in full non‐parametric maximum‐likelihood estimation of the two‐parameter logistic model

Parameters of the two‐parameter logistic model are generally estimated via the expectation–maximization (EM) algorithm by the maximum‐likelihood (ML) method. In so doing, it is beneficial to estimate the common prior distribution of the latent ability from data. Full non‐parametric ML (FNPML) estimation allows estimation of the latent distribution with maximum flexibility, as the distribution is modelled non‐parametrically on a number of (freely moving) support points. It is generally assumed that EM estimation of the two‐parameter logistic model is not influenced by initial values, but studies on this topic are unavailable. Therefore, the present study investigates the sensitivity to initial values in FNPML estimation. In contrast to the common assumption, initial values are found to have notable influence: for a standard convergence criterion, item discrimination and difficulty parameter estimates as well as item characteristic curve (ICC) recovery were influenced by initial values. For more stringent criteria, item parameter estimates were mainly influenced by the initial latent distribution, whilst ICC recovery was unaffected. The reason for this might be a flat surface of the log‐likelihood function, which would necessitate setting a sufficiently tight convergence criterion for accurate recovery of item parameters.     

A comparison of ranking and rating procedures for value system measurement

Compared the validity and reliability of 2 value measurement techniques. 296 Ss (161 females and 135 males) in introductory psychology filled out the 2 measurement techniques and an attitude survey. The Rokeach Value Survey instructed Ss to separately rank 2 sets of 18 values in order of importance. A rating version of the Value Survey instructed Ss to rate the same 36 values from 1 to 99. 236 Sreturned 6 weeks later and again filled out both measurement techniques. Results of the multimethod factor analysis indicate very good convergent validity among the 4 measures of a given value (2 techniques × 2 sessions) and very good discriminant validity between measures of different values. Probably due to the ipsative nature of the ranking procedure, the test-retest reliabilities were higher for the ranked measurements than for the rated measurements. The construct validity of both measurement techniques, as determined by multiple regression and analysis of variance, were similar. Despite criticisms of ranking procedures, both the ranked and the rated versions were of equal reliability and validity.     

Vibrotactile masking: A comparison of psychophysical procedures

The amounts of ipsilateral and contralateral masking obtained with a modified psychophysical procedure were compared for vibratory stimuli presented to different body loci. Results of a combined forced-choice localization task and a forced-choice detection task were similar to the results obtained in previous investigations which employed method of limits and two-alternative, temporal forced-choice procedures, respectively. Ipsilateral maskers produced similar amounts of masking in both the detection and localization tasks for fingertip and arm test sites. Contralateral maskers resulted in considerably more masking in the localization task than in the detection task for both fingertip and arm test sites. When large longitudinal distances were introduced between the test stimulus and masker, little masking was evident in either the localization or detection task. It was concluded that the differences in the amount of ipsilateral and contralateral masking obtained with different psychophysical procedures reflect different effects of a masker on the detectable attributes of a test stimulus. Implications of these results for the study of multiple tactile perception were discussed.     

Response elimination: A comparison of four procedures

Four procedures for eliminating an operant response were compared within a multiple schedule. Response reduction was most rapid when reinforcement was provided for a specific alternative response. The decline in responding produced by extinction and differential reinforcement of other behavior was similar. Responding initially increased and then gradually decreased when reinforcers were delivered independently of responding at fixed times. Removal of reinforcement from the alternative-response and DRO procedures did not result in recovery of the target response. However, the shift from response-independent dilivery of reinforcers to extinction caused an initial increase in responding.     

A comparison of adaptive psychometric procedures based on the theory of optimal experiments and bayesian techniques: implications for cochlear implant testing

Numerous previous studies have focused on the development of quick and efficient adaptive psychometric procedures. In psychophysics, there is often a model of the psychometric function supported by previous studies for the task of interest. The theory of optimal experiments provides a framework for utilizing a model of the process to develop quick and efficient sequential-testing strategies for estimating model parameters, making it appropriate for developing adaptive psychophysical-testing methods. In this study, we investigated the application of sequential parameter search strategies based on the theory of optimal experiments and Bayesian adaptive procedures for measuring psychophysical variables. The results presented in this article suggest that more sophisticated psychometric procedures can expedite the measurement of psychophysical variables. Such techniques for quickly collecting psychophysical data may be particularly useful in cochlear implant research, where a large set of psychophysical variables are useful for characterizing the performance of an implanted device. It is to be hoped that further development of these techniques will make psychophysical measurements available to clinicians for tuning and optimizing the speech processors of individual cochlear implant patients.     

Memory for targets in a multi‐level simulated‐environment: A comparison between able‐bodied and physically disabled children

Two groups of children, one able‐bodied and the other with physical disabilities, explored a symmetrical three‐tiered virtual building that contained six distinctive target objects, two on each storey. In a subsequent test, the target objects were removed and participants were asked to make judgments of the directions to the former target locations from each floor in turn. At each test site, judgments were required for targets that were formerly on the same floor and for those on higher and lower floors. Relative tilt error scores suggested a bias for both groups, in that targets that were higher than the test location were judged as consistently lower than their actual position, whereas targets that were lower than the test location were judged as higher than their actual position. Absolute tilt errors revealed an asymmetry in both groups, with more accurate tilt errors for judgments directed to lower than higher floors. The relevance of these results for the source of the asymmetry is discussed.     

Comparing the squared multiple correlation coefficients of non‐nested models: An examination of confidence intervals and hypothesis testing

The performance of the asymptotic method for comparing the squared multiple correlations of non‐nested models was investigated. Specifically, the increase in a given regression model's R² when one predictor is added was compared to the increase in the same model's R² when another predictor is added. This comparison can be used to determine predictor importance and is the basis for procedures such as Dominance Analysis. Results indicate that the asymptotic procedure provides the expected coverage rates for sample sizes of 200 or more, but in many cases much higher sample sizes are required to achieve adequate power. Guidelines and computations are provided for the determination of adequate sample sizes for hypothesis testing.