Ex post facto designs

They are characterized because the researcher cannot intentionally manipulate the VI or randomly assign subjects to its different levels. Subjects are not random. Subjects are selected based on whether or not they possess certain characteristics. Subjects are selected after VI has occurred. The relationships of "pre-existing" variables are studied.

The VI can be: Organismic: sex, age, personality characteristic, intelligence, anxiety, illness... etc. Outside the body: having lived through a catastrophe, belonging to an educational system, social environment... etc.

There are two research strategies, which will divide the designs following Leon Montero: Retrospective: The causal process has already occurred and it is a matter of looking for (reconstructing the facts) the possible causes that have caused them. Prospective: LV is known (not RV) but its consequences have not been evaluated.

Index

1. General characteristics
2. Control techniques
3. Use of ex post facto designs in applied research

I saw her: the values ​​are given. There is no intentional manipulation of this variable, but "value selection". It is not possible to establish causal relationships as in the experiment. Only the relationship between variables that covariate with the phenomenon studied can be studied. To establish causal relationships, 3 requirements must be met:

1. Existence of covariation between LV and RV.
2. The VI must precede the RV.
3. Relevant explanations must be able to be ruled out.

These designs only fulfill the first and sometimes the second, but not the third. For this reason, although we talk about LV and RV, the difference between both variables is merely theoretical. When the ex post facto design is for group comparison, hypotheses are formulated in which a differential relationship is established between the groups.

Kerlinger (1984) à In ex post facto research, inferences are made about the relationship between variables, without direct intervention, from the concomitant variation of VVII and VVDD. Although this design has less internal validity than the experimental one, it gains in external validity, since the investigations are They are usually carried out in natural situations and without more representative than the experiments also in terms of subjects and variables. It is very useful in the applied field, since it allows addressing issues that could not be investigated from the experimental method.

To establish the relationship between the variables, we must control the possible VVEE: Alvarado (2000) indicates three procedures:

• The pairing or mating of subjects: The use of subject variables and the impossibility of randomization. It consists of selecting for each group, subjects with equal values ​​in the most relevant VVEE (EX: if we are studying the relationship between depression-VI- and absenteeism work-DV- we can match the subjects in other variables related to work absenteeism, such as: level of education, level of anxiety, illnesses suffered, etc. Groups of subjects would be formed that had equal levels of these variables).
• Analysis of covariance (ANCOVA): control technique on VVEE, which requires the use of representative samples. Control through statistical procedures carried out after data collection. Eliminates the influence of the disturbing variable on RV, debugging the effect of IV on RV.
• Introduction of variables related to DV. Use multiple VVDDs instead of just one (EX. Absenteeism VI,, VVDD depression and job satisfaction)> Next: Part1: Classification of designs Ex post facto

Clinical psychology: to establish diagnostic categories or make predictions about diagnosis and therapy 50's to evaluate the effects of psychotherapy. Neurophysiological investigations: The functions of the hemispheres have been studied using simple prospective designs (in this type of Research subjects are selected according to the brain lesions they suffer and their behaviour).

Epidemiological investigations: study of health and disease in human populations (includes disease, quality of health, and mental health).

• Descriptive studies: The survey methodology is used and is carried out when little is known about the occurrence, natural or specific history of a disease. Estimate the frequency or trend of a disease in a particular population and generate specific etiological hypotheses.
• Etiological studies: when the disease is well known and specific hypotheses are available. Identify risk factors for the disease, estimate their effects on the disease, and suggest possible intervention strategies.
• Educational scope: in the study of variables related to school performance or success (self-concept, sex, aptitude, cultural or racial differences, etc.). Developmental psychology: investigations that try to establish the relationship between age and the use of learning strategies, memory, etc.

This article is merely informative, in Psychology-Online we do not have the power to make a diagnosis or recommend a treatment. We invite you to go to a psychologist to treat your particular case.

If you want to read more articles similar to Ex post facto designs - Meaning and control techniques, we recommend that you enter our category of Experimental psychology.