An Improved Rank Correlation Effect Size Statistic for Single-Case Designs: Baseline Corrected Tau

Measuring treatment effects when an individual’s pretreatment performance is improving poses a challenge for single-case experimental designs. It may be difficult to determine whether improvement is due to the treatment or due to the preexisting baseline trend. Tau-U is a popular single-case effect size statistic that purports to control for baseline trend. However, despite its strengths, Tau-U has substantial limitations: Its values are inflated and not bound between −1 and +1, it cannot be visually graphed, and its relatively weak method of trend control leads to unacceptable levels of Type I error wherein ineffective treatments appear effective. An improved effect size statistic based on rank correlation and robust regression, Baseline Corrected Tau, is proposed and field-tested with both published and simulated single-case time series. A web-based calculator for Baseline Corrected Tau is also introduced for use by single-case investigators.

[1]  Rumen Manolov,et al.  A comparison of mean phase difference and generalized least squares for analyzing single-case data. , 2013, Journal of school psychology.

[2]  Benjamin G. Solomon,et al.  Violations of Assumptions in School-Based Single-Case Data , 2014, Behavior modification.

[3]  G. Huston.  The Hospital Anxiety and Depression Scale. , 1987, The Journal of rheumatology.

[4]  Ann Casey,et al.  A Methodology for the Quantitative Synthesis of Intra-Subject Design Research , 1985 .

[5]  Yanni G. Malliaris,et al.  Rumination-focused cognitive behaviour therapy for residual depression: a case series. , 2007, Behaviour research and therapy.

[6]  Kimberly J. Vannest,et al.  The Theil–Sen Slope for High-Stakes Decisions from Progress Monitoring , 2012 .

[7]  H. B. Mann,et al.  On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other , 1947 .

[8]  Robin K. Morgan,et al.  Single-participant research design. Bringing science to managed care. , 2001, The American psychologist.

[9]  B. Gorman,et al.  Design and Analysis of Single-Case Research , 1996 .

[10]  B. Gorman,et al.  Calculating effect sizes for meta-analysis: the case of the single case. , 1993, Behaviour research and therapy.

[11]  Petra Kaufmann,et al.  Experimental And Quasi Experimental Designs For Research , 2016 .

[12]  Allison Gorman Faith,et al.  Meta-analysis of Single-Case Research , 2014 .

[13]  R. Wilcox.  A Note on the Theil-Sen Regression Estimator When the Regressor Is Random and the Error Term Is Heteroscedastic , 1998 .

[14]  Richard I. Parker,et al.  Robust regression for single-case data analysis: How can it help? , 2011, Behavior research methods.

[15]  M. Sidman,et al.  Tactics of scientific research : evaluating experimental data in psychology , 1961 .

[16]  M. Kendall.  A NEW MEASURE OF RANK CORRELATION , 1938 .

[17]  Jonathan M. Campbell.  Statistical Comparison of Four Effect Sizes for Single-Subject Designs , 2004, Behavior modification.

[18]  Bruce A. Thyer,et al.  Single-case experimental designs: Strategies for studying behavior change (2nd edn): by David H. Barlow and Michel Hersen Pergamon Press, New York, 1984, 419 pp. , 1985 .

[19]  Rumen Manolov,et al.  Estimating Slope and Level Change in N = 1 Designs , 2010, Behavior modification.

[20]  Richard I. Parker,et al.  Controlling Baseline Trend in Single-Case Research. , 2006 .

[21]  P. Sen.  Estimates of the Regression Coefficient Based on Kendall's Tau , 1968 .

[22]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[23]  Jacob Cohen.  Statistical Power Analysis for the Behavioral Sciences , 1969, The SAGE Encyclopedia of Research Design.

[24]  David J. A. Dozois,et al.  A psychometric evaluation of the Beck Depression inventory-II , 1998 .

[25]  F. Wilcoxon.  Individual Comparisons by Ranking Methods , 1945 .

[26]  Sheng Yue,et al.  The influence of autocorrelation on the ability to detect trend in hydrological series , 2002 .

[27]  Wayne F. Velicer,et al.  The Reliability and Accuracy of Time Series Model Identification , 1983 .

[28]  J. McKean,et al.  Design Specification Issues in Time-Series Intervention Models , 2000 .

[29]  Alan E. Kazdin,et al.  Single-Case Research Designs: Methods for Clinical and Applied Settings , 2010 .

[30]  Idit Keidar,et al.  Introduction , 2019, Concurrency: the Works of Leslie Lamport.

[31]  W. Shadish,et al.  The state of the science in the meta-analysis of single-case experimental designs , 2008 .

[32]  D. Barlow,et al.  Single Case Experimental Designs: Strategies for Studying Behavior Change , 1976 .

[33]  D. Watson,et al.  Development and validation of brief measures of positive and negative affect: the PANAS scales. , 1988, Journal of personality and social psychology.

[34]  A. Beck,et al.  Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation , 1988 .

[35]  W. Shadish.  Analysis and meta-analysis of single-case designs: an introduction. , 2014, Journal of school psychology.

[36]  Bradley E. Huitema,et al.  Autocorrelation in applied behavior analysis: A myth. , 1985 .

[37]  A. Beck,et al.  An inventory for measuring depression. , 1961, Archives of general psychiatry.

[38]  Justin D. Smith,et al.  Single-case experimental designs: a systematic review of published research and current standards. , 2012, Psychological methods.

[39]  Victor L. Willson,et al.  Modeling Causal Error Structures in Longitudinal Panel Data: A Monte Carlo Study , 2000 .

[40]  Kimberly J. Vannest,et al.  Combining nonoverlap and trend for single-case research: Tau-U. , 2011, Behavior therapy.

[41]  Kimberly J. Vannest,et al.  An improved effect size for single-case research: nonoverlap of all pairs. , 2009, Behavior therapy.

[42]  A. Siegel.  Robust regression using repeated medians , 1982 .

[43]  Kimberly J. Vannest,et al.  Effect Sizes in Single Case Research: How Large is Large? , 2005 .

[44]  Erin E. Barton,et al.  Comparison of Overlap Methods for Quantitatively Synthesizing Single-Subject Data , 2010 .

[45]  Gregory E. Granato.  Kendall-Theil Robust Line (KTRLine--version 1.0)-A Visual Basic Program for Calculating and Graphing Robust Nonparametric Estimates of Linear-Regression Coefficients Between Two Continuous Variables , 2006 .

[46]  Rumen Manolov,et al.  Percentage of nonoverlapping corrected data , 2009, Behavior research methods.

[47]  H. Theil.  A Rank-Invariant Method of Linear and Polynomial Regression Analysis , 1992 .

[48]  B. Thorn,et al.  Evidence-based practice in psychology. , 2006, The American psychologist.

[49]  Rumen Manolov,et al.  Comparing N = 1 Effect Size Indices in Presence of Autocorrelation , 2008, Behavior modification.

[50]  Joseph W. McKean,et al.  Irrelevant autocorrelation in least-squares intervention models. , 1998 .

[51]  R. Senf.  Tactics of Scientific Research: Evaluating Experimental Data in Psychology , 1962 .

[52]  Richard I. Parker,et al.  The Relationship Between Visual Analysis and Five Statistical Analyses in a Simple AB Single-Case Research Design , 2006, Behavior modification.

[53]  B. E. Huitema.  Autocorrelation: 10 Years of confusion , 1988 .

[54]  Kimberly J. Vannest,et al.  Effect Size in Single-Case Research: A Review of Nine Nonoverlap Techniques , 2011, Behavior modification.

[55]  J. Levin,et al.  Single-case research design and analysis : new directions for psychology and education , 1992 .

[56]  W. Shadish,et al.  Characteristics of single-case designs used to assess intervention effects in 2008 , 2011, Behavior research methods.

[57]  P. Young,et al.  Time series analysis, forecasting and control , 1972, IEEE Transactions on Automatic Control.