Pre-experimental plans and design of the "true" psychological...

Pre-experimental plans and design of "true" psychological research

Pre-experimental design. The most famous authors in the field of experimental psychology, psychodiagnostics and qualitative evaluation in general, whose work we often refer to, are foreign authors - J. Campbell and R. Gottsdanker, A. Anastasi and others, as well as domestic psychologists - A. A. Bodalev, V. N. Druzhinin, LF Burlachuk, T. V. Kornilova, A. G. Shmelev et al.

During the presentation of the "design experimental schemes of research, we are inclined to apply the notation set forth in the editorship of J. Campbell. The X the experimental variable or event is designated, the influence of which on the subjects is to be measured. The variable X is called an independent variable (NEP). The O Is the object of investigation, as well as the measurement procedure and the results of the psychological measurement. Edit About under the influence of the X is called the dependent variable (ZP). X and "O", standing in one line, refer to the same specific persons. The direction from left to right indicates the time order of application and measurement of variables.

We will use this symbolism in the process of presenting the structure and features of any experimental plan.

Case study (plan 1). During the design of the experiment, the psychologist selects the optimal study plan (s) in accordance with the experimental hypothesis and research conditions. The simplest scheme of the experiment is the so-called case study (Plan 1). In a symbolic form, this plan looks like "X O". This design is defined as the procedure for the one-time measurement of psychological variables ("O") after the stimulus (independent variable) ("X") is applied to one experimental sample.

Experiments on plan 1 are conducted, for example, in the professional selection of personnel for the purpose of "aerobatic" research to determine an indicative set of psychodiagnostic methods for the study of any of the qualities through a valid instrument. The results of this type of research are realized, in the first case, in the development of a test battery (this is an example of the correct application of plan 1 in practice), and in the second - by evaluating psychological qualities (example "pseudo experiment"),

Unfortunately, some studies, especially those conducted for the purpose of "training", are built according to a plan that provides for a single examination in one group that has undergone some psychological or organizational and management impact. Such studies completely lack control over the parameters of the experimental situation, and they do not represent scientific value. For learning purposes, such a plan is only applicable as part of familiarizing students with the methodology or technology of research.

The application of this scheme of research in practice, as a rule, looks like this. A survey of the subjects is carried out, and the results of the description of the applied test are interpreted. Such a study does not take into account the fact that the technique was developed in other subjects and in other situations (the situation of "convenient sampling"), and therefore it is unlikely that the validity indicated in the description of the test actually takes place. If to a theoretical validity this situation relates in part, then it is not necessary to speak about empirical. The psychologist should imagine that each technique should be practically re-designed and developed on that particular population with which he works. This process of revalidating a psychological measuring tool was called adaptation of the method. For such situations, J. Campbell pointed out that "the existence of absolute knowledge or the knowledge of the essence of isolated objects is an illusion."

However, this practice of testing exists. Especially it is common among beginning psychologists and poorly versed in the substance of the case customers of psychological research. Recently, the situation is gradually beginning to improve, and it is even considered unethical to provide research to protect the research that has been carried out under plan 1. Indeed, in this scheme of research one can not even reason approximately about what is the impact of the "X" led to changes in the investigated variable and it is this impact that is measured by testing.

One thing to remember is that plan 1 is the starting point in considering psychological research schemes.

Plan with preliminary and final testing on one group (plan 2). Design of pre-experimental research with preliminary and final testing on one group - is often carried out in pedagogical development and vocational training.

An analysis of the validity control of the psychological experiment conducted in Plan 2 shows the possibility of preliminary psychodiagnostics of the subjects during the design of the methodological means of investigation (battery tests, testing and adaptation of techniques, etc.) for the purposes of personnel selection, -pedagogical tasks.

Comparison of static groups (plan 3). The design of this psychological study looks like:

As follows from the scheme, two heterogeneous groups participate in the study. The experimental sample is affected by an independent variable X and the measurement of the dependent variable - the final testing of the "About" - is carried out in the situation of absence of preliminary psychological examination. The control group undergoes simultaneous measurement with the experimental group.

The results of the study of the internal validity of the experiment according to plan 3 show that the quotient "Von controlled, since the psychological dimension was carried out after the impact once, the factor "Natural development" controlled due to single testing after exposure, factors Test Effect and Reactive effect are also monitored in the absence of repeated (preliminary) testing. The Dropout completely controlled, as there is no drop-out. Factor Instrumental error It is not controlled due to the uncertainty of the conclusion about the validity of the tests. Factor Statistical regression is not monitored due to lack of selection and the ability to make an unambiguous conclusion about including the best and worst subjects. Factors Testing and Interaction of selection factor with natural development are not controlled, as there is no selection.

Sources of external non-existence - factors "Reaction of subjects for experiment", "Interaction of selection factor and experimental impact", "Interference interference (imposition) of experimental effects" are not controlled.

Psychological experiments on plan 3 in the course of research are more preferable than other pre-experimental schemes, since they assume the presence of a control group. This type of design and the factors influencing the validity of the study are subject to greater control by psychologists.

Pre-experimental plans are widely used in the course of research on the professional suitability of specialists at enterprises, in solving problems of rational placement of personnel. Schemes of research on plans 1-3 are the "reference point" in the classical experimental work of psychologists and are actively used by them in practice. The possibilities for controlling the factors of invalidity are limited, but according to the results of pre-experimental "sketches" the researcher already can present some psychological tendencies and more thoroughly prepare a true experiment. Therefore, pre-experimental schemes are actively used in the so-called aerobatic studies.

Psychological research is based on the practical application of the so-called "true" plans. Experiment, when control groups are used during the study. Schemes of experiments of this kind are designated as plans 4, 5 and 6.

Design true experiment.

Plan with preliminary and final testing and control group (plan 4). Scheme 4 is a classic "design" psychological experimental research. Its feature is not only the presence of a control group (it is already present in the pre-experimental scheme 3), but also the equivalence of the experimental and control samples.

The choice of the experiment plan with the preliminary and final testing and the control group is made in accordance with the experimental task and the research conditions. When it is possible to form at least two homogeneous groups, the following scheme is used:

The R means that the homogeneity of the groups is obtained by randomization.

For practical assimilation of the possibilities of implementing the experimental plan 4, it is possible to conduct a study of the laboratory forming experiment, which contains a mechanism for confirming the hypothesis that positive motivation affects the concentration of human attention.

Experiment Procedure:

1. Formation of experimental and control samples. The study participants are divided into pairs, carefully equalized by the indicators of preliminary testing or by variables that are significantly correlated with each other. Then the members of each pair are randomly by lot are included in the experimental or control groups.

2. Both groups are asked to test the test "Correction test with rings (O1 and O3).

3. The activity of the experimental sample is stimulated. Suppose that the test subject is given an experimental stimulus (X).

4. Both groups are asked to test the " Correction test with syllables (O2 and O4).

Analysis of the results of the experiment.

1. Empirical data are tested for normal distribution.

2. The calculation of the arithmetic mean (Mx) and standard deviation (δχ) of the results of preliminary and final tests is performed.

3. The average values ​​of test scores in the experimental and control groups (O1 ↔ O3, O1 ↔ O2, O3 ↔ O4, O2 ↔ O4) are compared.

4. A comparison is made of the mean values ​​for the Student's t-test, i.e. determination of the statistical significance of the mean differences.

5. The following relations are proved: O1 = O3; O1 & lt; O2; O3 = O4 and O2 & gt; O4, as indicators of the effectiveness of the experiment.

6. A study is made of the validity of the experiment by determining the degree of control of factors of invalidity.

In the theory of psychological research, there are two categorically opposite opinions to the problem of external validity.

On the one hand, D. Hume's assertion that "induction or generalization can never be completely logically justified" is widespread. If internal validity can be defined within the logic of probability statistics, then the problems of external validity are not completely resolved in any clear and convincing way. If we carry out research on a certain set of subjects, then the probability of an error in spreading the results to another sample is very high.

On the other hand, many scholars share the postulate of JS Mill on the laws of nature - "glueiness" ("stickiness") of nature, which states that the closer two events in space and time, as well as the measured values ​​for some or all parameters, the more they will follow the same laws.

In conclusion, it should be added that, as J. Campbell pointed out, the optimal method of determining the effect of research may be still "common sense" and "considerations of a non-mathematical nature", however "a good experimental plan is independent of the way statistical processing of results".

Plan R. Solomon for the four groups (plan 5). If the tasks and research conditions allow for the formation of four equivalent samples, then the experiment is constructed according to plan 5, named after its compiler - R. Solomon .

Solomon's plan is an attempt to compensate for factors that threaten the external validity of the study. To the elements of plan 4, two groups (experimental and control) are added, in which preliminary testing is not performed.

Comparison of data on additional groups neutralizes the impact of testing and the impact of the study environment itself, and also allows for better generalization of the results.

The identification of the effect of the experimental effect is reproduced by proving the following inequalities: O2 & gt; O1; O2 & gt; O4; O5 & gt; O6; O3 & gt; O3.

If all four relationships are satisfied, then the legitimacy of the output significantly increases.

The use of plan 5 determines the likelihood of neutralizing the interaction between testing and experimental impact, which facilitates the interpretation of the results of studies under plan 4.

Comparison of O6 with O1 and O3 reveals a combined effect of natural development and background. Comparison of the mean O2 and O5; O4 and O6 allow to estimate the main effect of preliminary testing. Comparison of the mean O2 and O4; O5 and O6 - the main effect of the experimental effect.

If the main effect of preliminary testing and the interaction effect are small and they can be neglected, then it is desirable to perform the covariance analysis of O4 and O2, using the results of preliminary testing as a companion variable.

A plan with a control group and testing only after exposure (plan 6). Very often when performing experimental tasks, researchers face the situation of the need to study psychological variables in the conditions of the impossibility of conducting a preliminary examination of subjects, since the research is conducted after the impact of independent variables, when the event has already occurred and it is necessary to identify its consequences. Here the optimal test plan is a control group and testing plan only after exposure. Using randomisation, homogeneous experimental and control samples of the subjects exposed to the independent variable are formed. The testing of variables is carried out after the experimental action:

Thus, the basis of the experimental method in psychology is the so-called true plans in which the control, in fact, of all the main factors affecting internal validity is controlled (Table 5.3). The reliability of the results in the experiments planned in Schemes 4-6 is beyond doubt among a significant majority of researchers. The main problem, as in all other psychological studies, is the search and application of adequate measuring aids.

Table 5.3

Analysis of the validity of the experiments on the plans 4-6

List of sources of internal non-existence of the experiment

Plan 4

Plan 5

Plan 6

Background

+

+

+

Natural development

+

+

+

Test Effect

+

+

+

Instrumental error

+

+

+

Statistical regression

+

+

+

Composition of the group

+

+

+

Shootout

+

+

+

The interaction of the composition of the group with natural development, etc.

+

+

+

Test interaction and X

?

The list of sources of external non-existence of the experiment

Reaction of subjects to experiment

?

?

?

The interaction of the composition of groups and X

?

?

?

Interaction between different X

The notion of nesting. Nsing (from English, nesting) is a collection of homogeneous objects inside larger units. This is an option for a two-level hierarchical classification signs. Any analytical work, including psychological, includes different methods. Nesting appears as a result of the holistic influence of an independent variable consisting of several interconnected parts (x1 + x2 + ... + xn) under conditions of analysis of this complex stimulus.

Assume that when conducting a study of the effectiveness of the new organization of enterprises, the conditions for implementing the management methodology of the firm by various directors with different professional experience were studied - the following factors occurred: the experience of the director and the methods of managing the firm. These factors overlap, i.e. their interaction is revealed.

If you take into account the third factor - the "gender of the leader", the "nesting" appears. Since the factors experience & quot ;, the gender (male/female) and the organization management method can be represented as dichotomous, then the "two-level nesting" is formed. The fragmentation of the investigated factors into more hierarchies leads to the formation of a more complex "hierarchical classification" or complex nesting. Analysis of psychological information, built in the form of nesting, is most often carried out according to the scheme shown in Table 5.4.

Table 5.4

Scheme two-level classification signs (nesting)

Method 1

Method 2

Man

Great experience

Small Experience

Woman

Great experience

Small Experience

Empirical data on various items can be compared among themselves by applying various statistical communication measures and mathematical criteria.

For example, during the analysis of the socio-demographic characteristics of the personnel of a computer firm, a mixed view of the hierarchical classification (nesting) was obtained (Table 5.5).

Table 5.5

Data analysis of the socio-demographic characteristics of the company's personnel (in %)

Age

Sex

Marital status

Family

Idle

Up to 24 years old -

11 people (27.5%)

9

91

Men

45

0

100

Women

55

17

83

25-30 years -

11 people (27.5%)

55

45

Men

36

50

50

Women

64

57

43

31-35 years - four people (10%)

75

25

Men

50

100

0

Women

50

50

50

36-40 years - eight people (20%)

88

12

Men

25

50

50

Women

75

100

0

41 years and older - six people (15%)

100

0

Men

67

100

0

Women

33

100

0

It can be seen from the table that the two characteristics of the subjects (gender and marital status) have a two-level hierarchical structure, and the third factor (age) is a five-level structure.

As an example of a nesting interpretation, let us analyze the indicator, which reveals the fact that 57% is the total proportion of women aged 25-30-years old (64%) who are married and have families. Moreover, this age group of subjects in itself is 27.5% of the entire experimental sample.

Nesting, as a classification, allows you to conveniently present the experimental data for analysis, visual demonstration and publication.

Also We Can Offer!

Other services that we offer

If you don’t see the necessary subject, paper type, or topic in our list of available services and examples, don’t worry! We have a number of other academic disciplines to suit the needs of anyone who visits this website looking for help.

How to ...

We made your life easier with putting together a big number of articles and guidelines on how to plan and write different types of assignments (Essay, Research Paper, Dissertation etc)