Elsevier

Behavior Therapy

Volume 48, Issue 1, January 2017, Pages 115-127
Behavior Therapy

Methodological Paper
Analyzing Therapeutic Change Using Modified Brinley Plots: History, Construction, and Interpretation

https://doi.org/10.1016/j.beth.2016.09.002Get rights and content

Highlights

  • Idiographic information about individual therapy outcome is important to development of effective therapies.

  • Conventional nomothetic, group-based analyses of research do not make such information readily available.

  • Modified Brinley plots are an idiographic analysis technique.

  • Reliable change and clinical cutoffs permit determination of clinically significant change and deterioration.

  • Group data such as means, confidence intervals, and effect sizes can be shown on modified Brinley plots.

Abstract

The paper reviews the history, construction, and interpretation of modified Brinley plots, a scatter plot used in therapy outcome research to compare each individual participant’s scores on the same dependent variable at Time 1 (normally pretreatment baseline; x-axis), with scores at selected times during or after treatment (y-axis). Since 1965 eponymously named Brinley plots have occasionally been used in experimental psychology to display group mean data. Between 1979 and 1995 a number of clinical researchers modified Brinley plots to show individuals’ data but these plots have received little subsequent use. When constructed with orthogonal axes having the same origin and scale values, little or no change over time is shown by individuals’ data points lying on or closely about the diagonal (450o) while the magnitude and direction of any improvement (or deterioration), outliers, and the extent of replication across cases shows via dispersion of points away from 450o. Interpretation is aided by displaying reliable change boundaries, clinical cutoffs, means, variances, confidence intervals, and effect sizes directly on the graph. Modified Brinley plots are directly informative about individual change during therapy in the context of concurrent change in others in the same (or a different) condition, clearly show if outcomes are replicated and if they are clinically significant, and make nomothetic group information, notably effect sizes, directly available. They usefully complement other forms of analysis in therapy outcome research.

Section snippets

Brinley Plots: A Short History

The history begins with a type of scatter plot developed by Brinley (1965), who measured the performance speed of young and elderly participants in several different cognitive tasks and calculated the group mean speed for each task separately for younger (My) and older (Mo) participants. He then plotted points representing each My (x-axis) and same-task Mo (y-axis) pair. With the origin and scales of the scatter plot axes the same, if there were no systematic performance differences between

Creation and Interpretation of Modified Brinley Plots

Therapy outcome research typically involves participants sharing a common problem who are repeatedly observed (up to z times), normally at t1 before therapy and then again later (t2Z; t2 hereafter, for simplicity). Many protocol variations are possible, with participants divided into control as well as therapy groups and with repeated measures for all participants in baseline and during and after intervention, and measurement of multiple dependent variables (DVs). The net result is a matrix of

Displaying Means, CIs, and ESs

To this point the paper has shown how modified Brinley plots permit the informative display of each individual’s data and the immediate perception of their reliable and clinically significant change. It next discusses how nomothetic information can be overlaid on the plot, focusing in particular on means, CIs, and various ES measures.

Between-Group Comparisons

So far the focus has been on examining within-participant change over time. Typical therapy outcome studies, however, often combine within- and between-participant conditions, commonly having both therapy and control/comparison groups. Modified Brinley plots are first and foremost a within-subject data analysis tool, but that does not prevent them from being used to compare change over time in different groups. In Jacobson et al. (1984) and Sobell et al. (1995), data from different therapy

Utility and Limitations of Modified Brinley Plots

In contrast to conventional graphs reporting group data, modified Brinley plots primarily display individuals’ data and are, therefore, idiographic—they drill down from the group aggregate level to display patterns of individual response to therapy (Sobell et al., 1995). They do not, however, necessarily identify who each data point represents, and this is a limitation in the extent to which they are fully idiographic. Can these plots be made even more idiographic, perhaps by displaying

Conclusion

Cohen (1990) reflected that, in changing psychology’s methodological practices, “Things take time” (p. 1311). Even so, Stunkard and Penick’s (1979) inaugural modified Brinley plot did not deserve to suffer more than three decades of obscurity and neglect, despite attempts by Jacobson et al. (1984) and Sobell and colleagues (1995) to revive interest. Neither has their concern that nomothetic research does not deliver information directly useful to clinicians gone away in the meantime. Recently,

Conflict of Interest Statement

The author declares no conflicts of interest in the production of this paper.

References (89)

  • D.H. Barlow et al.

    Why can’t we be more idiographic in our research?

    Perspectives on Psychological Science

    (2009)
  • A.E. Bergin et al.

    Changing frontiers in the science of psychotherapy

    (1972)
  • N.M. Blampied

    Single-case research adaptations for the analysis of group data

  • N.M. Blampied

    Single-case research designs and the scientist-practitioner ideal in applied psychology

  • N.M. Blampied

    Adapting single-case research designs for research with groups

    (2016)
  • J.M. Blennerhassett et al.

    The test-retest reliability and responsiveness to change of the Hand Function Survey during stroke rehabilitation

    Australian Occupational Therapy Journal

    (2010)
  • J.J. Borckardt et al.

    Time-series statistical analysis of single case data

  • J.C. Bourret et al.

    Visual analysis in single-case research

  • J.F. Brinley

    Cognitive sets, speed and accuracy of performance in the elderly

  • S.K. Card et al.

    Readings in information visualization: Using vision to think

    (1999)
  • W.S. Cleveland

    Graphs in scientific publications

    American Statistician

    (1984)
  • W.S. Cleveland

    Visualizing data

    (1993)
  • J. Cohen

    Things I have learned so far

    American Psychologist

    (1990)
  • J. Cohen

    The earth is round (p < .05)

    American Psychologist

    (1994)
  • J.O. Cooper et al.

    Applied behavior analysis

    (2007)
  • G. Cumming

    Understanding the new statistics: Effect sizes, confidence intervals, and meta-analysis

    (2012)
  • G. Cumming

    The new statistics: Why and how

    Psychological Science

    (2014)
  • P.J. Curran et al.

    Twelve frequently asked questions about growth curve modelling

    Journal of Cognition and Development

    (2010)
  • R. Dar et al.

    Misuse of statistical tests in three decades of psychotherapy research

    Journal of Consulting and Clinical Psychology

    (1994)
  • D.H. Davis et al.

    Augmenting visual analysis in single-case research with hierarchical linear modelling

    Behavior Modification

    (2012)
  • W.G. Dye et al.

    Increasing speed of processing with action video games

    Current Directions in Psychological Science

    (2009)
  • S. Few

    Now you see it: Simple visualization techniques for quantitative analysis

    (2009)
  • F. Fidler et al.

    Toward improved statistical reporting in the Journal of Consulting and Clinical Psychology

    Journal of Consulting and Clinical Psychology

    (2005)
  • W.W. Fisher et al.

    Visual aids and structured criteria for improving visual inspection and interpretation of single-case designs

    Journal of Applied Behavior Analysis

    (2003)
  • C.O. Fritz et al.

    Effect size estimates: Current use, calculations and interpretations

    Journal of Experimental Psychology: General

    (2012)
  • H.G. Funkhouser

    Historical development of the graphical representation of statistical data

    Osiris

    (1937)
  • G. Gigerenzer et al.

    Surrogate science: The idol of a universal method for scientific inference

    Journal of Management

    (2015)
  • H.A. Gordon et al.

    Clinically significant symptom reduction in children with attention-deficit/hyperactivity disorder treated with micronutrients: An open label reversal design study

    Journal of Child and Adolescent Psychopharmacology

    (2015)
  • W.H. Gottdiener

    Improving the relationship between the randomized clinical trial and real-world clinical practice

    Psychotherapy

    (2011)
  • S.N. Hayes et al.

    The idiographic application of nomothetic empirically based treatments

    Clinical Psychology: Science and Practice

    (1999)
  • R. Hubbard

    Alphabet soup: Blurring the distinctions between p’s and α’s in psychological research

    Theory and Psychology

    (2004)
  • R. Hubbard et al.

    Why p values are not a useful measure of evidence in statistical significance testing

    Theory and Psychology

    (2008)
  • N.S. Jacobson et al.

    Statistics for assessing the clinical significance of psychotherapy techniques: Issues, problems, and new developments

    Behavioral Assessment

    (1988)
  • N.S. Jacobson et al.

    Clinical significance: A statistical approach to defining meaningful change in psychotherapy research

    Journal of Consulting and Clinical Psychology

    (1991)
  • I gratefully acknowledge Heather Duncan, Brian Haig, Joanna Lothian, Jess McIvor, Cathy Robson, Julia Rucklidge, and Ellen Sole for their help in the preparation of this paper.

    View full text