What is it like to be a dependent variable?

CiNaPS Conference: “Causality in the Neuro- and Psychological Sciences”
29-30 October 2018, Antwerp, Belgium.

What is it like to be a dependent variable?

Abstract
In psychotherapeutic research, the ‘randomized controlled trial’ is held as gold standard to reach causal statements on treatment efficacy.[i]In this interventionist design, the therapeutic intervention is considered the independent variable and is either contrasted to a no-treatment control or an alternative treatment parallel condition.[ii]Iff all possibly influencing factors are kept constant, the difference in interventions is thought to causea measured or observed difference in the dependent variable, which is the amount of symptoms shown by patients, the receivers of treatment. Within this design, it is thus assumed that the human beings who participate in the study are the carriers of specified variables.[iii]Critical literature is primarily focused on validity of instruments, demarcation and measurement of variables, epistemic issues in randomization and circumstances of treatment, et cetera, yet little attention is given to the most vital predisposition of this design, which is the participant. As it is his or her collected ‘data’ that allow for an aggregated comparison of outcomes, this paper is focused on the process of data retrieval from patients who follow treatment in a research context. What is it like for patient-participants to be considered a ‘dependent variable’, and how may that affect the object under investigation, the therapeutic intervention?

In this paper, we discuss a case who voluntarily participated in a randomized controlled efficacy study that was focused on psychotherapy for major depression.[iv]The patient, a 47-year old male, was randomly assigned to a psychotherapist for a 20-session weekly depression treatment. Before, during and after treatment, he was interviewed by ‘his own’ researcher. Narrative from therapy andresearch were analyzed with an interpretative phenomenological approach[v], to understand how the patient experienced the ‘role’ or ‘function’ of being a participant in a scientific study, and how this may have impacted his therapeutic process.

The selected case is a critical case, which is not necessarily representative for the larger population, but is highly informative towards the attribution of causality based on the strictly defined methodological procedure in psychotherapy research.[vi]By discussion of the idiosyncrasy of experience and the perceived impact of the research procedure onto the data gathered during the therapeutic intervention, we question whether it is justified to aggregate data from this idiosyncratic case into group data, which is a vital element to the concept of causality within the gold standard method. The question is not how causality works for our individual patient, as it is assumed that such individual processes would level out via randomization on group level. Rather, the question is whether the causal attribution within the rationale of an interventionist design remains valid when the object of study appears to bealtered by the act of studying it.

[i]Wampold, B. E. (2001). The great psychotherapy debate. Models, methods and findings. New York: Routledge;
cf. Cartwright, N. (2010). What are randomised controlled trials good for? Philosophical Studies, 147, 59–70.

[ii]Kendler, K. S., & Campbell, J. (2009). Interventionist causal models in psychiatry: repositioning the mind–body problem. Psychological Medicine, 39, 881–887.

[iii]Cf. Woodward, J. (2011). Data and phenomena: a restatement and defense. Synthese, 182, 165-179.

[iv]Meganck, R., Desmet, M., Bockting, C., Inslegers, R. Truijens, F., De Smet, M., et al. (2017). The Ghent Psychotherapy Study (GPS) on the differential efficacy of supportive-expressive and cognitive behavioral interventions in dependent and self-critical depressive patients: study protocol for a randomized controlled trial. BioMed Central, 18:126. doi.: 10.1186/s13063-017-1867-x.

[v]Smith, J. A., & Eatough, V. (2007). Interpretive Phenomenological Analysis. In E. Lyons & A. Coyle (eds). Analysing qualitative data in psychology. London: Sage.

[vi]Cf. Truijens, F. L. (2016). Do the numbers speak for themselves? A critical analysis of procedural objectivity in psychotherapeutic efficacy research. Synthese, 194, 4721-4740.

 

Advertisements

Comments are closed.