The longitudinal method of research in psychology , as a rule, is opposed to the analytical model of slices. Recently, it has come to be seen in the context of identifying experimental delayed effects. Let us further consider what constitutes a longitudinal research method .
General information
The longitudinal method consists in repeatedly fixing the parameters on one person or group of people. The slicing model, in contrast to it, involves a comparison of indicators at the same time for representatives of different age categories. The classical longitudinal method in psychology means "continued study."
Specificity
The longitudinal comparative method is in a special position in the structure of analytical technology, social sciences, and disciplines that study behavior. This is due to a number of circumstances. First of all, a special situation is connected with the specificity of the tested development hypotheses. Equally important are the difficulties of planning, organizing observations, and processing results. Many authors gave classifications to the applied analysis models in their works. The considered longitudinal method, according to Ananiev, relates , in particular, to organizational methods.
Structural elements
Development hypotheses contain an assumption about the dynamics of changes in indicators over time. However, this factor is not considered as a source or prerequisite. It is regarded as an analogue of an independent variable. The theoretical justification of the possibility of temporal dynamics of changes in indicators is interpreted as development, it also provides methodological principles for understanding this process, the provisions of a specific concept, as well as an assessment of observation planning.
Problem solving
The longitudinal method allows you to directly turn to checking casual assumptions in terms of requirements for the time sequence of consequences and causes. Accordingly, it can bring to the implementation of two key conditions for identifying communication. The first involves the study of causes and consequences in time, the second - the establishment of covariance between them. The place of prerequisites can be occupied by any effects that are under observation. However, they cannot be interpreted as experimental if the specialist does not manage them. Other requirements of the conclusion about the causes can be obtained in sequential cross-sectional or slice observations. For example, the condition for the presence of covariance between variables is detected by means of intergroup differences or non-zero correlations between the variables. The requirement for the absence of alternative justifications can be implemented through the use of statistical or experimental control.
Development features
The longitudinal method arose during the introduction of a systematic census in Quebec, Canada in the 17th century. This analytical model was most developed after the First World War in America. Subsequently, at the end of the 20th century. longitudinal method entrenched in social disciplines and the science of behavior. The modern development of the model is determined by the improvement of information analysis techniques determined at the stage of observation planning. The authors of one of the articles devoted to the method indicate that in most modern theories, statements that have a dynamic character are made indirectly or directly. In other words, they appeal to the justification of a particular phenomenon in the context of changes occurring with it or its connections with other phenomena. A similar conclusion can be made regarding the psychological patterns that are established when testing hypotheses about the development, delayed or long-term effects of the effects.
Relationship with Empirical Observations
Hypothesis testing is a key task that the longitudinal method performs. However, despite this, development conclusions are often made in accordance with empirical observations. They are carried out in the framework of various psychological concepts using the slice method. It allows you to detect the relationship of several static variables taken in a separate time period. The use of the findings is determined by the existence of an unspoken assumption about the equivalence of the samples through which the comparison is carried out, as well as historical periods for various categories of subjects. This often leads to the neglect of an important source of confusion, which must be given special attention.
Key concepts
To indicate the community of people in the sample by year of birth, a term such as "cohort" is used. In accordance with the demographic characteristics, this concept means a certain group of people, designated within a geographic or other population, having experienced similar events in a given time period. The age variable represents the chronological number of years at the time of observation. The analysis should also clarify the concept of "period". It denotes the time of measurement and the stage that encompasses the life of the cohort, including historical events common to its members. Formally, community is defined as follows:
Cohort = Period of measurement (calendar. Year) - age (number of years from birth).
Explanations
The above equation illustrates a linear relationship between measurement time, cohort and age. In this case, the source of systematic mixing, important for the longitudinal method, is expressed. People born in the same year live in general social conditions covering a specific historical period. From this follows such a conclusion. The cohort will be common to people not only the year of birth, but also their “history” - the content of the period in which they live in a particular country, in specific geographical conditions, political, economic, cultural space. If this confusion is ignored, then the validity of the conclusions obtained by the specialist using the longitudinal method can be called into question.
The consequences
Linear dependence leads to the fact that during the control of any two indicators, the third variable is also controlled. If the study uses the slicing method, the sample of people also has a common "history", but it differs for participants in the longitudinal section and sections. This leads to a confusion of the factor of social circumstances and age. In this regard, when performing cross-sectional comparisons of the parameters of people of different ages, the revealed differences between the more mature and younger subjects can express not the line of development of the main process, but the effects of the cohort. The use of the longitudinal method with multiple consecutive measurements can contribute to the discovery of results not specified as the subject of research, but the effects of social circumstances, as a historical stage specific to a given sample.
Attempts to Overcome Addiction
They are divided into 2 conceptual categories. The first is Mason's research. It is supposed to solve the problem at the statistical level. For this, models are formed by means of which collinearity (absolute mathematical dependence) between the cohort, age and time period is eliminated. The second group contains approaches that suggest a theoretical justification for the process of excluding the consideration of the impact of one indicator on the identified lines of development or their rethinking. In this direction, several techniques have been developed. Some consider cohort parameters as an interaction of age and time effects. Others replace the sample with its characteristics, which can be precisely defined and measured. In the ideal case, the effects of the period and the cohort, which have a fundamentally different explanatory status than the time indicators, are excluded from the analysis. They will be replaced by operationalized properties that make it possible to breed parameters of age, historical period, and the sample itself. This form of analysis is fundamentally impossible outside the framework of a “true” longitudinal study, where many measurements are carried out for several cohorts simultaneously.
Goals
The longitudinal method allows one to test the "strong" casual hypotheses when performing a quantitative assessment of the dynamic properties of development. The key objectives of the study are:
- Improving the accuracy of measuring the effect. It is achieved by controlling intraindividual variability. In this case, repetitive observation schemes are used, which, among others, include the longitudinal method.
- Testing hypotheses related to the orientation of casual relationships, assessing their strength.
- Determination of the functional form of development curves or intraindividual trajectories.
- Analysis of inter-individual differences. It is carried out using casual models.
In the literature, the lack of consensus on the issue of the minimum number of time slices is a key difference in the understanding of the considered method.