Question

A psychologist compared the effectiveness of Cognitive Behavioural Therapy (CBT) and eye movement desensitisation and reprocessing (EMDR) for the treatment of post-traumatic stress disorder (PTSD Clients with PTSD were recruited and randomised into either the CBT or EMDR group. A group of 5 psychologists were trained in both treatment methods. Individual clients were treated by the same psychologist across the study. The Clinician-Administered PTSD Scale (CAPS) was used to assess severity of PTSD symptoms before and after the course of therapy. They then analysed the difference in CAP Scale. The predictor variables were psycholo- gist, and treatment method. (a) What analysis should be used for these data? Explain why. Write the statistical model for this as an equation (b) Could one of the predictor variables be thought of as a block variable in this case? Why?

Answer 1

Before answering the question ,let us try to understand as to what is the problem scenario.

It is an experimental design in which two sample of patients with PTSD are exposed to two treatment types - CBT and EMDR .

This treatment is done by any of the five e physhologists who have competency of treatment of both methods.

In this case, thus, the variables which may impact the outcome/ cure /improvement in PTSD status are :_

1. The selection of a Psychologist

2. Selection of the treatment methodology

Hence we notice that the outcome (the CAP Score ) is dependent on two independent variables also called as predictor variables since they predict the outcome of the CAP Score , and which are

1. The selection of a Psychologist

2. Selection of the treatment methodology.

The mutual dependencies between all the given variables (independent and dependent) is analysed and described by a statistical tool known as regression analysis.

Linear Regression Analysis is a statistical tool which established a direct relationship between all the variables involved in the experiment/observation along with their level of significance of impact.

It consists of 3 stages –

(1) analyzing the correlation and directionality of the data,

(2) estimating the model, i.e., fitting the line, and

(3) evaluating the validity and usefulness of the model.

Hence due to completeness of the analysis ,we select regression .

It also tells us if the variable are :-

1. significant

2. extent of dispersion

3. level of correlation between variables

4. Level of data variablity explained by the regression equation.

5. the confidence interval of the variables.

Thus in the given case the statistical tool to use to analysis is regression analysis and the analytical model is given by

$CAP =\alpha+\beta_{1}*(Psychologist-type)+\beta_{2}*(treatment-method)$

Where the variables Psychologist -type and treatment _method denote

1. The selection of a Psychologist

2. Selection of the treatment methodology. respectively.

b). Checking if one of the independent variables can be a block variables

Let us firstly try to nuderstand as to what is a blocking variable ?

A categorization variable for information within a dataset that is used to control, test, or manipulate the distribution and statistical results. These variables should be observed, reliable, and unchanging. For example, a dataset can be blocked based on name, date, race or gender. It's a way of viewing the data from a preferred reference point within a random collection of information.

Clearly ,the psychoogists can be a block variable based on names of the psychologists.