Lecture
3 types of control:
- control associated with the possibilities of planning as the construction of experimental schemes;
- control involving the specification of different types of variables and, accordingly, different types of experiments;
- control involving the need to eliminate threats to a valid inference posed by other factors.
1. Control by means of experimental schemes.
Planning an experiment is the organization of its conduct in accordance with the assumed experimental model, through a mental comparison with a flawless model of its realization, in order to obtain reliable data.
Planning presupposes that in the EH the IV and DV are made concrete and, at the methodological level, operationalized.
A distinction is drawn between formal and substantive planning.
Formal planning is possible in those studies in which the principle traditional for the natural sciences is used: the variables (the IV and additional variables) are represented and managed independently of one another. If there is more than one managed variable, then the principle of isolated conditions is adopted. The principle of isolated conditions: the levels of each variable may be combined with the levels of the other variables, and from the relationships of the individual variables one can construct their overall effect – the result of the action of the IV and the interaction effects.
Formal planning is aimed at the choice of a scheme, a plan for organizing the interactions, and the specification of the minimal effect (in the differences in the DV).
The experimental plan (design) includes the specification of the scheme for presenting the IV conditions to different groups of participants or the sequence of IV levels presented to a single participant, as well as the number of trials or participants (n). It is also the plan for recording the DV.
Substantive planning – at the stage of formulating the hypotheses, the specificity of the psychological reality under study must be preserved, made concrete in the basic processes singled out for investigation. Here the type of variables is determined from the standpoint of the specified hypothetical constructs and the resolution of questions about the operationalization of the variables, along with the choice of methodological means.
To achieve validity, various means of experimental control are used.
It presupposes methods of setting the levels of the IV (functional control of the IV, management of experimental factors) and control of all other variables as sources of competing explanations, i.e. factors that may threaten a valid inference about the action of the IV on the DV. The principal (but not the only) component of experimental control is the development of experimental schemes or plans.
Experimental control is a condition for control over the inference.
2. Control of the type of variables.
Determining the type of variables from the standpoint of their correspondence to real conditions or to the theoretical constructs under consideration makes it possible to predict the possibility of generalizing conclusions beyond the experimental situation. That is, here there are certain requirements for external validity. The method of selecting participants into groups or from the population determines the permissible limits of generalization. Depending on the status of particular variables in relation to the EH and the possibilities of managing them, such forms of control as stabilization of variables or their maximal variation may be used.
3. Primary control and the experimenter effect.
Primary control is the control of all extraneous variables, primarily those connected with the isolation of the IV and the interaction between the experimenter and the participant, carried out prior to the choice of experimental plan and independently of it.
The interaction of the researcher with the participant, motivation; the influence of the personality traits of the experimenter and the participant. Rosenthal: the «Pygmalion effect» – the experimenter’s involuntary facilitation of the manifestation of precisely those regularities in whose truth he is convinced (i.e. leading the participant toward the desired strategy).
Comments