Confidence intervals for the parameters of psychometric functions

A Monte Carlo method for computing the bias and standard deviation of estimates of the parameters of a psychometric function such as the Weibull/Quick is described. The method, based on Efron's parametric bootstrap, can also be used to estimate confidence intervals for these parameters. The method's ability to predict bias, standard deviation, and confidence intervals is evaluated in two ways. First, its predictions are compared to the outcomes of Monte Carlo simulations of psychophysical experiments. Second, its predicted confidence intervals were compared with the actual variability of human observers in a psychophysical task. Computer programs implementing the method are available from the author.     

Confidence intervals for standardized linear contrasts of means

Most psychology journals now require authors to report a sample value of effect size along with hypothesis testing results. The sample effect size value can be misleading because it contains sampling error. Authors often incorrectly interpret the sample effect size as if it were the population effect size. A simple solution to this problem is to report a confidence interval for the population value of the effect size. Standardized linear contrasts of means are useful measures of effect size in a wide variety of research applications. New confidence intervals for standardized linear contrasts of means are developed and may be applied to between-subjects designs, within-subjects designs, or mixed designs. The proposed confidence interval methods are easy to compute, do not require equal population variances, and perform better than the currently available methods when the population variances are not equal.     

Confidence intervals for gamma-family measures of ordinal association

This research focused on confidence intervals (CIs) for 10 measures of monotonic association between ordinal variables. Standard errors (SEs) were also reviewed because more than 1 formula was available per index. For 5 indices, an element of the formula used to compute an SE is given that is apparently new. CIs computed with different SEs were compared in simulations with small samples (N = 25, 50, 75, or 100) for variables with 4 or 5 categories. With N > 25, many CIs performed well. Performance was best for consistent CIs due to N. Cliff and colleagues (N. Cliff, 1996; N. Cliff & V. Charlin, 1991; J. D. Long & N. Cliff, 1997). CIs for Spearman's rank correlation were also examined: Parameter coverage was erratic and sometimes egregiously underestimated.     

Confidence intervals for the overall effect size in random-effects meta-analysis

One of the main objectives in meta-analysis is to estimate the overall effect size by calculating a confidence interval (CI). The usual procedure consists of assuming a standard normal distribution and a sampling variance defined as the inverse of the sum of the estimated weights of the effect sizes. But this procedure does not take into account the uncertainty due to the fact that the heterogeneity variance (tau2) and the within-study variances have to be estimated, leading to CIs that are too narrow with the consequence that the actual coverage probability is smaller than the nominal confidence level. In this article, the performances of 3 alternatives to the standard CI procedure are examined under a random-effects model and 8 different tau2 estimators to estimate the weights: the t distribution CI, the weighted variance CI (with an improved variance), and the quantile approximation method (recently proposed). The results of a Monte Carlo simulation showed that the weighted variance CI outperformed the other methods regardless of the tau2 estimator, the value of tau2, the number of studies, and the sample size.     

Searching for conscientiousness on the MMPI-2

It has been argued that the revised Minnesota Multiphasic Personality Inventory (MMPI-2) fails to measure the Big Five dimension of conscientiousness (C) an assertion supported by the research finding that none of the factors underlying the MMPI correlate substantially with the C factor of the NEO Personality Inventory (NEO-PI)-a popular operationalization of the Big Five model. In this study we reconsidered the relationship between the MMPI-2 item pool and the C factor of the NEO-PI-R (Form S). Collections of MMPI-2 C markers did not correlate much higher than .5 with the NEO-PI-R factor scale in cross-validation hold-out samples. Most pervasive, however, was the finding that MMPI-2 item markers of C were also strongly (in many cases, more highly) related to the NEO-PI-R factor of neuroticism (N). Efforts to purify collections of MMPI-2 C markers, first by careful item selection and then by suppression of N variance, were met with limited success. Intercorrelations among NEO-PI-R scales suggest that the difficulty in finding markers that discriminate between N and C extends beyond any shortcomings of the MMPI-2 item pool.     

Searching for Conscientiousness on the MMPI-2

It has been argued that the revised Minnesota Multiphasic Personality Inventory (MMPI-2) fails to measure the Big Five dimension of conscientiousness (C) an assertion supported by the research finding that none of the factors underlying the MMPI correlate substantially with the C factor of the NEO Personality Inventory (NEO-PI)—a popular operationalization of the Big Five model. In this study we reconsidered the relationship between the MMPI-2 item pool and the C factor of the NEO-PI-R (Form S). Collections of MMPI-2 C markers did not correlate much higher than .5 with the NEO-PI-R factor scale in cross-validation hold-out samples. Most pervasive, however, was the finding that MMPI-2 item markers of C were also strongly (in many cases, more highly) related to the NEO-PI-R factor of neuroticism (N). Efforts to purify collections of MMPI-2 C markers, first by careful item selection and then by suppression of N variance, were met with limited success. Intercorrelations among NEO-PI-R scales suggest that the difficulty in finding markers that discriminate between N and C extends beyond any shortcomings of the MMPI-2 item pool.     

Use of the MMPI-2 in American Indians: I. Comparability of the MMPI-2 between two tribes and with the MMPI-2 normative group

The comparability of the MMPI-2 in American Indians with the MMPI-2 normative group was investigated in a sample of 535 Southwestern and 297 Plains American Indian tribal members with contrasting sociocultural and historical origins. Both American Indian tribal groups had clinically significant higher T scores (> 5 T points) on 5 validity and clinical scales, 6 content scales, and 2 supplementary scales than did the MMPI-2 normative group. There were no significant differences between the 2 tribal groups on any of the MMPI-2 clinical, content, or supplementary scales. Matching members of both tribes with persons in the MMPI-2 normative group on the basis of age, gender, and education reduced the magnitude of the differences between the 2 groups on all of these scales, although the differences in T scores still exceeded 5 T points. It appears likely that the MMPI-2 differences of these 2 American Indian groups from the normative group may reflect their adverse historical, social, and economic conditions.     

A Structural Analysis of the Impact of MMPI-2 on MMPI-1

Minnesota Multiphasic Personality Inventory-1 (MMPI-1) items modified for Minnesota Multiphasic Personality Inventory-2 (MMPI-2) are presented in pairs to illustrate modifications. MMPI-1 items deleted from MMPI-2 are presented, grouped by categories. Frequencies of items remaining in MMPI-2 for the following special scales are presented: Harris and Lingoes subscales of the clinical scales (Harris & Lingoes, 1955/1968); Wiggins Content scales (Wiggins, 1966); Tryon, Stein, and Chu (TSC) cluster scales (Stein, 1968; Tryon, 1966); Indiana Rational scales (Levitt, 1989); and selected other special scales. Frequencies of MMPI-1 items in the MMPI-2 validity and content scales are also presented, k is concluded that most of the special scales developed for MMPI-1 remain relatively intact in MMPI-2.     

Assessing general maladjustment with the MMPI-2

The validities of 7 MMPI-2 (Butcher, Graham, Ben-Porath, Tellegen, & Kaemmer, 2001) measures of general maladjustment were compared using a composite criterion measure based on self-reported symptom severity and clinicians' ratings of symptom severity and level of functioning. Participants were 274 male and 425 female clients at a community mental health center and 105 male and 247 female clients at a university psychological clinic. All MMPI-2 measures were significantly related to the composite criterion measure for both male and female clients in both settings. The mean score on 8 clinical scales (M8) consistently was the best indicator of maladjustment. Although other MMPI-2 measures sometimes added significantly to the variance accounted for in the criterion measure, increments were small and probably not clinically meaningful. However, M8 added significantly and meaningfully to each of the other MMPI-2 measures in predicting maladjustment. Implications for using the MMPI-2 to assess general maladjustment in outpatient mental health settings are discussed.     

A shortened version of the MMPI-2

A psychometrically sound method of prorating scores from a shortened version of the MMPI-2 is presented to approximate the full-scale raw scores on the basic scales. After a brief review of the history of short versions of the original MMPI and their strengths and weaknesses, justifications for developing and publishing this new version are offered. In spite of the risk of abuse by harassed and over-worked clinicians, there are cogent reasons to make this set of procedures available to practitioners and research investigators. These procedures were devised on the data from the 2,600 men and women in the original MMPI-2 restandardization sample and cross-validated on a sample of 632 test records from a psychiatric inpatient service. The dependability of estimated single raw scores as well as of the patterning of the prorated profile patterns is explored.     

Diagnostic construct validity of MMPI-2 Restructured Form (MMPI-2-RF) scale scores

In the current investigation, the authors examined the diagnostic construct validity of the Minnesota Multiphasic Personality Inventory-2-Restructured Form (MMPI-2-RF) in a patient sample. All participants were diagnosed via the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I/P). The data set used in this study was composed of 544 patients--67 with bipolar disorder, 407 with major depressive disorder, and 70 with schizophrenia. Multivariate analyses revealed a pattern of mean scale differences among patient groups that was mostly consistent with the prominent features of each diagnostic group; logistic regression analyses identified a number of scales that were strong, unique predictors in the differentiation between pairs of diagnostic groups. The higher order scales (H-O)--the Emotional/Internalizing Dysfunction (EID) and Thought Dysfunction (THD) scales were most useful in differentiating between patient groups. For differentiating bipolar disorder patients from the other diagnostic groups, the Activation (ACT) Specific Problem scale was most useful. Although not all hypothesized scale differences emerged; overall, the pattern of results provides support for the diagnostic construct validity of the MMPI-2-RF scales.     

Use of the MMPI-2 With Chicanos: Strategies for Counselors

The Minnesota Multiphasic Personality Inventory–2 (MMPI-2) is frequently used by professional counselors to assess Chicano or Mexican American clients in various mental health settings. It is important to understand how ethnic and sociocultural variables (e.g., ethnic identity, acculturation, or racism) may influence the assessment process. This article presents strategies for the culturally competent use of the MMPI-2 with this burgeoning Hispanic subpopulation.     

An approach for practice administration and interpretation of the MMPI-2

This paper discusses one teaching approach for practice administration and interpretation of the MMPI-2 in a graduate course on personality assessment. After graduate students practice scoring and interpreting the MMPI-2 inventory completed with specific response sets and practice interpreting MMPI-2 profiles of cases prepared for teaching and practice in interpretation, students practice administering and interpreting the MMPI-2 for volunteer test-takers who complete the inventory with a simulated stress response set. The simulated stress response set approach may be helpful in providing skill-building practice in the early phases of learning the MMPI-2 with volunteer test-takers, while minimizing some of the ethical concerns raised regarding actual administrations with nonclient volunteers in the early phases of training in assessment.     

Random responding to the MMPI-2 F scale

Exact probabilities for a variety of F scale cutoffs are presented for detecting examinees who respond randomly to the MMPI-2 items.