Lecture
Population validity is the aspect of extending a generalization beyond the experimental situation to those populations from which the participants of the study were selected.
Population hypotheses differ from experimental ones; they are tested in other types of research – in sample surveys. They are not causal and are formulated as statements about the distribution of one or another property in a population (given the possibility of measuring it in the individuals of a representative sample).
A special problem here is taking into account the equal accessibility of the population. It is also necessary to obtain everyone's consent.
More often, researchers deal with situations in which two unknown populations are compared by contrasting the sample values of the variable under study. «Unknown» means there are no prior data on how the property of interest is distributed in the population, or what inter-individual differences can be expected.
The research goal is usually the comparison of populations (through the comparison of sample means?).
Diagnostic variables characterizing personality are singled out, and the similarity/difference between the samples representing the populations is established.
Sometimes such surveys are called statistical, because instead of experimental control, confoundings with extraneous factors are here isolated statistically.
Sample surveys can also be called those in which a certain intervention is present, and its effect is assessed in relation to the entire population that underwent the intervention. That is, here an assumption is added about an influencing factor as the cause of the differences.
But these are pre-experimental plans, for example, where the intervention is a new method of teaching and the like.
Important concepts:
The general population is the totality of units (people) about which it is necessary to draw a conclusion on the basis of a sample.
The sample is a part (subset) of the general population.
The sample and the population are connected through the selection procedure.
Representativeness of the sample is the main aspect of assessing population validity, which presupposes that, as a result of using special strategies for selecting subjects from a broader population, the sample of participants shown in the study well represents this population of people.
Population validity is an assessment of how legitimate it is to extend the generalizations obtained on the basis of using a limited number of subjects in the study to broader samples of subjects, to populations.
The problem of estimating a psychological variable acting as the dependent variable, when its sample values are given.
At first the population is known – then it is unknown (the example with the Unified State Exam).
The goal is often the establishment of the distributions of psychological attributes.
In sample surveys, special attention must be paid to extraneous factors that lead to confoundings with the factor being studied.
Example: A. Prikhozhan's study: it examined the property of anxiety in adolescents in regions affected by the Chernobyl disaster.
The application of psychodiagnostic techniques.
Problem: confounding with the task factor (differences in the construction of the image of the task).
SAMPLES: probability samples (simple, systematic, stratified – in 2 stages: distribution across strata and probability selection); selective (targeted) – for example, cluster.
Comments