Cross-sectional time-series experiments: Some suggested statistical analyses.

In the past, statistical analyses for time-series experiments have usually operated with a single-case model, thereby limiting the general applicability of the designs. In this article, alternative analytical procedures are developed for cross-sectional time-series in which the sample size is large and the number of observations per case is relatively small. Interrupted time series, equivalent time samples, and multiple time series are all treated within a multiple regression framework. A generalized least squares estimation procedure is outlined as a more suitable alternative to the G. E. Box and G. M. Jenkins (1970) approach. Some of the special advantages of the designs are briefly discussed. (22 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Erratum to Simonton.

Reports an error in the original article ("Cross-Sectional Time-Series Experiments: Some Suggested Statistical Analyses") by Dean Keith Simonton (Psychological Bulletin, 1977[May], Vol 84[3], 489-502). On page 491, the second line of Footnote 1 reads: "chap. 3; Overall & Klett, 1972, Pt. 5). Here, dummy. . . ." It should read: "variables in multiple regression, including dummy. . . ." On page 492, Equation 2 reads: "Yt = b?+ b?Xt + b?t + b?Xt." It should read: "Yt = b?Xt + b?t + b?Xtt." On page 493, the second sentence in Paragraph 2 reads: "Since t = 0 at the moment of intervention, b? is the expected mean value of Yt across all cases just prior to the intervention." It should read: "Since t = 0 at the moment of intervention, b? is the expected mean value of Yt across all cases just prior to the intervention." On page 495, the first sentence in Paragraph 5 reads: "Coefficient b? estimates an interaction effect, namely the difference between the intercept of the postintervention control group (i.e., b?) and the intercept of the postintervention experimental group." It should read: "Coefficient b? estimates an interaction effect, namely the difference between the intercept of the postintervention control group (i.e., b? + b?) and the intercept of the postintervention experimental group." (The following abstract of this article originally appeared in record 1978-00178-001.) In the past, statistical analyses for time-series experiments have usually operated with a single-case model, thereby limiting the general applicability of the designs. In this article, alternative analytical procedures are developed for cross-sectional time-series in which the sample size is large and the number of observations per case is relatively small. Interrupted time series, equivalent time samples, and multiple time series are all treated within a multiple regression framework. A generalized least squares estimation procedure is outlined as a more suitable alternative to the G. E. Box and G. M. Jenkins (1970) approach. Some of the special advantages of the designs are briefly discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Measuring the effect of changing legislation on the frequency of divorce: The Netherlands, 1830-1990

This article discusses different procedures for measuring the effects of judicial changes on divorce rates. It presents an alternative model and applies it to a historical time series for the Netherlands. In this model, intervention variables were added to a statistical time-series (ARIMA) model. The conclusion of our analysis was that the effects of three judicial changes were only temporary.     

Interrupted time-series analysis with brief single-subject data.

Assessing change with short time-series data is difficult because visual inference is unreliable with such data, and current statistical procedures cannot control Type 1 error because they underestimate positive autocorrelation. This article describes these problems and shows how they can be solved with a new interrupted time-series analysis procedure (ITSACORR) that uses a more accurate estimate of autocorrelation. Monte Carlo analyses show that, with short series, ITSACORR provides better control of Type 1 error than all previous procedures and has acceptable power. Clinical examples also show that ITSACORR is easy to use and functions well with real data. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Comments on Rawlings' nonorthogonal analysis of variance.

In J. Overall and D. Spiegel's reply to R. Rawlings's (see record 1972-26084-001) criticism of their previous article, the authors state that Rawlings's alternative nonorthogonal analysis of variance is equivalent to their method, which Rawlings criticized as incorrect. In 2 separate articles (a) Rawlings replies to Overall and Spiegel's present article, and (b) I. Smith contends that there is a statistical error in G. Joe's (see record 1971-25969-001) attempt to clarify the original Overall and Speigel article. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Regression analyses of repeated measures data in cognitive research.

Repeated measures designs involving nonorthogonal variables are being used with increasing frequency in cognitive psychology. Researchers usually analyze the data from such designs inappropriately, probably because the designs are not discussed in standard textbooks on regression. Two commonly used approaches to analyzing repeated measures designs are considered in this article. It is argued that both approaches use inappropriate error terms for testing the effects of independent variables. A more appropriate analysis is presented, and two alternative computational procedures for the analysis are illustrated. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Alternatives to Simonton's analyses of the interrupted and multiple-group time-series designs.

D. K. Simonton (see record 1978-00178-001) discussed the use of several linear models for the analysis of data arising in the interrupted time-series design and the multiple-group time-series design. The present authors contend that the objectives of Simonton's analyses can be realized using profile analysis. Statistical procedures for analyzing the interrupted time-series and the multiple-group time-series designs are outlined. The procedures are applicable when several Ss are observed on several pre- and posttreatment occasions and when the number of Ss is greater than the number of occasions. (11 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A note on concurrent and predictive validity designs: A critical reanalysis.

Comments on an article in which G. Barrett et al (see record 1981-11631-001) argued that the conceptual distinction between predictive and concurrent validity has been exaggerated. The present authors offer a more complete classification of validity designs and illustrate sources of differences in the quality of the validity estimates obtained. Five kinds of predictive designs for selection research are identified: (a) follow-up with random selection, (b) follow-up with selection by existing systems, (c) selection based on test to be validated, (d) selection preceding testing, and (e) shelf research. For some of these, and for corresponding concurrent designs, problems arise from the lack of data for informed estimates of appropriate population parameters. It is pointed out that severe range restriction intensifies parameter estimation and the risk of Type II error. Corrections suggested for the effects of contaminations such as age, tenure, and experience are shown to be oversimplifications. It is concluded that on both conceptual and practical bases different validity designs are not equivalent or interchangeable. (10 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Irrelevant autocorrelation in least-squares intervention models.

Autocorrelation coefficients computed on the entire series of observations obtained from interrupted time-series designs are generally irrelevant to the general linear model (GLM) independence assumption. Consequently, the argument that GLM solutions are invalidated when large coefficients of this type are encountered is incorrect. Several decompositions of the terms involved in such coefficients are provided to show how deterministic components in the correct model contaminate these coefficients. Example data sets from articles in the methodological literature that were written to promote the use of complex time-series methods are used to illustrate relevant and irrelevant autocorrelations and to demonstrate the application and viability of GLM time-series intervention models. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

An approach for widening the bioequivalence acceptance limits in the case of highly variable drugs

PURPOSE: Highly variable drugs pose a problem in bioequivalence assessment because they often fail to meet current regulatory acceptance criteria for average bioequivalence (80-125%). This paper examines alternative approaches to establishing bioequivalence. METHODS: Suggested solutions have included alternate study designs, e.g., replicate and multiple dose studies, reducing the level of the confidence interval, and widening the acceptance limits. We focus on the latter approach. RESULTS: A rationale is presented for defining wider acceptance limits for highly variable drugs. Two previously described methods are evaluated, and a new method having more desirable properties is proposed. CONCLUSIONS: We challenge the "one size fits all" current definition of bioequivalence acceptance limits for highly variable drugs, proposing alternative limits or "goal posts" which vary in accordance with the intrasubject variability of the reference product.     

Pupillographic assessment of sleepiness in sleep-deprived healthy subjects

Medical research frequently involves the statistical comparison of >2 groups, often using data obtained through the application of complex experimental designs. Fortunately, inferential statistical methodologies exist to address these situations. Analysis of variance (ANOVA) in its many forms is used to simultaneously test the equality of all groups in a study. One-way (with 1 independent variable), 2-way (with 2 independent variables), and repeated-measures (patients serve as their own controls) ANOVAs are forms of this technique. Each form has been developed to analyze data from a specific experimental design. Analysis of covariance (ANCOVA) allows the researcher to control for confounding variables that may influence the response of the dependent variable. Finally, multivariate analysis of variance (MANOVA) evaluates the simultaneous responses of multiple dependent variables to > or = 1 independent variable. Whereas ANOVA is the correct alternative to statistically inappropriate multiple t-tests, MANOVA is the correct alternative to statistically inappropriate multiple univariate ANOVA calculations. Use of each of these statistical methods requires an appropriate experimental design and data meeting a number of assumptions. When used properly, each of these methods provides a powerful statistical analysis technique.     

Correction to Stephan and Rosenfield.

Reports an error in the original article by Walter G. Stephan and David Rosenfield (Journal of Personality and Social Psychology, 1978[Aug], Vol 36[8], pp. 795-804). Two corrections are needed to correlations in Table 2 on page 800. (The following abstract of this article originally appeared in record 1980-02037-001.) Examined the determinants of changes in the racial attitudes of 65 White elementary school children during school desegregation. A multiple regression analysis showed that increases in children's self-esteem, increases in children's interethnic contact, low parental authoritarianism, and nonpunitive parental child-rearing practices were all significantly related to positive changes in racial attitudes. In addition, data are presented that indicate that parental opposition to integration is an indirect determinant of change in children's racial attitudes because of its impact on children's interethnic contact. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Comment on Simonton's "Cross-sectional time-series experiments: Some suggested statistical analyses."

Suggests that the equation on which multiple time-series analysis is based is applicable to more than experimental and control groups that are identical prior to intervention as D. K. Simonton indicated (see record 1978-00178-001). The present article suggests that the equation is also applicable to cases in which comparable though not identical groups are investigated and in which nonequivalent groups are analyzed. (1 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Use of multiple regression in counseling psychology research: A flexible data-analytic strategy.

The purpose of this article is to explain the basic principles of multiple regression, to demonstrate the flexibility of this procedure for analyzing data from a number of designs, and to discuss issues related to the interpretation of results from a multiple regression analysis. The regression equations for a single independent variable and for two or more independent variables are presented, followed by a discussion of several coefficients related to these equations. Three analytic strategies are compared. Finally, several variations of the multiple regression theme, including the analysis of categorical variables, interactions, and covariance, are illustrated with an example from counseling research. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Stemming the TIDE: A correction of Fortner (1999) and a clarification of Larson and Hoyt (2007).

In this article, the author corrects an error in his 1999 dissertation (see record 2000-95004-391), which was summarized in detail shortly after its approval and was intensely examined more recently by D. G. Larson and W. T. Hoyt (see record #200711559-003). An error in the text of the dissertation transposed the variables of the denominator of the treatment-induced deterioration statistic (TIDE). The error was limited to the text of the document, and all calculations and results reported in the dissertation are correct. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

"Generalized single-case randomization tests: Flexible analyses for a variety of situations": Erratum.

Reports an error in "Generalized single-case randomization tests: Flexible analyses for a variety of situations" by Joel R. Levin and Bruce E. Wampold (School Psychology Quarterly, 1999[Spr], Vol 14[1], 59-93). (1) In Table 5 (p. 76), in the cell associated with the third row and third column, the upper value of 18.3 should be 18.73. (2) On p. 81, in the section titled "Simultaneous Start-Point Randomization for Matched Pairs of Units," the sentence beginning in line 10 should read: "With k? acceptable start points for the first pair and k? for the second, there would be k?k? summed sums and null hypothesis-compatible summed differences for the randomization distribution to test Rgen and Rcomp, respectively (as indicated in the Appendix, p. 89)." (The following abstract of the original article appeared in record 1999-10944-004.) A general class of single-case statistical procedures, derived from previously developed nonparametric randomization tests, is presented. These procedures are widely applicable in behavioral and educational research contexts in which only a few experimental "units" (individuals, small groups, or classrooms) are the recipients of one or more experimental treatments or interventions. Specifically illustrated are designs that focus on both the general and comparative effectiveness of alternative interventions (including control/placebo treatments), multiple units with differentiable characteristics (i.e., intervention by unit characteristic interactions), and multiple outcome measures (i.e., intervention by outcome measure interactions). Also provided are operational modifications that enhance the internal validity of studies incorporating single-case randomization-based analyses. Appropriate cautions and issues surrounding the use of these procedures are included. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Correction to Greville and Buehner (2010).

Reports an error in "Temporal predictability facilitates causal learning" by W. James Greville and Marc J. Buehner (Journal of Experimental Psychology: General, 2010[Nov], Vol 139[4], 756-771). Figure 2 (p. 759) contained an error. The corrected figure appears in the correction. (The following abstract of the original article appeared in record 2010-22538-005.) Temporal predictability refers to the regularity or consistency of the time interval separating events. When encountering repeated instances of causes and effects, we also experience multiple cause–effect temporal intervals. Where this interval is constant it becomes possible to predict when the effect will follow from the cause. In contrast, interval variability entails unpredictability. Three experiments investigated the extent to which temporal predictability contributes to the inductive processes of human causal learning. The authors demonstrated that (a) causal relations with fixed temporal intervals are consistently judged as stronger than those with variable temporal intervals, (b) that causal judgments decline as a function of temporal uncertainty, and (c) that this effect remains undiminished with increased learning time. The results therefore clearly indicate that temporal predictability facilitates causal discovery. The authors considered the implications of their findings for various theoretical perspectives, including associative learning theory, the attribution shift hypothesis, and causal structure models. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Therapeutic assessment for preadolescent boys with oppositional defiant disorder: A replicated single-case time-series design.

The Therapeutic Assessment (TA) model is a relatively new treatment approach that fuses assessment and psychotherapy. The study examines the efficacy of this model with preadolescent boys with oppositional defiant disorder and their families. A replicated single-case time-series design with daily measures is used to assess the effects of TA and to track the process of change as it unfolds. All 3 families benefitted from participation in TA across multiple domains of functioning, but the way in which change unfolded was unique for each family. These findings are substantiated by the Behavior Assessment System for Children (Reynolds & Kamphaus, 2004). The TA model is shown to be an effective treatment for preadolescent boys with oppositional defiant disorder and their families. Further, the time-series design of this study illustrated how this empirically grounded case-based methodology reveals when and how change unfolds during treatment in a way that is usually not possible with other research designs. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A note concerning reflection effects versus framing effects.

A recent article discussing the costs and benefits of judgment errors and presenting a taxonomy of judgment behaviors (H. R. Arkes; see record 1992-04093-001) included an error. This note corrects the error in Arkes's article in which framing effects were mislabeled as reflection effects. In addition, readers may have mistakenly received the impression that evidence for reflection and framing effects is more consistent than it is. So, it is clarified that research evidence for reflection and framing effects is variable rather than uniformly positive. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Errata to Carlson and Timm.

Reports an error in the original article by J. E. Carlson and N. H. Timm (Psychological Bulletin, 1974[Sep], Vol No.[81], 563-570). Errors made in some equations and estimations of the analysis of variance parameters are corrected. (The following abstract of this article originally appeared in record 1975-02163-001.) Discusses nonorthogonal fixed-effects experimental designs using both the full-rank and reduction in error sums of squares conceptualizations of data analysis. The hypotheses tested by several commonly used methods of analysis are clarified, and suggestions for choice of the most appropriate procedure are proposed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)