The longitudinal method of research in psychology is usually opposed to the analytical model of slices. Recently, it has been considered in the context of revealing experimental delayed effects. Let us further consider what the longitudinal research method is.
General information
Longitudinal method consists in multiple fixation of parameters on one person or group of people. The slice model, in contrast, involves comparing indicators for the same time among representatives of different age categories. The classic longitudinal method in psychology means "continued study".
Specifics
Longitudinal comparative method has a special position in the structure of analytical technique, social sciences, behavioral disciplines. This is due to a number of circumstances. First of all, a special position is associated with the specifics of the tested hypotheses about development. Of no small importance are the difficulties of planning, organizing observations, and processing results. Many authors gave classifications in their works to the applied analysis models. Considered longitudinalmethod, according to Ananiev, refers, in particular, to organizational techniques.
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 the independent variable. The theoretical substantiation of the possibility of temporal dynamics of changes in indicators is interpreted as a development, it also provides for methodological principles for understanding this process, the provisions of a specific concept, as well as an assessment of monitoring planning.
Problem solving
The longitudinal method allows you to directly address the verification of causal assumptions in terms of requirements for the temporal sequence of effects and causes. Accordingly, it can bring two key conditions for link detection closer to realization. The first involves the study of cause and effect in time, the second - the establishment of covariance between them. Prerequisites can be replaced by any effects that are under observation. At the same time, they cannot be interpreted as experimental if a specialist does not control them. Other causal inference requirements can be obtained from serial cross-sectional or cross-sectional observations. For example, the condition for the presence of covariance between variables is revealed through intergroup differences or non-zero correlations between 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 originated with the introduction of a systematic population census in Quebec in 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. the longitudinal method has taken root in the social sciences and behavioral sciences. The modern development of the model is determined by the improvement of information analysis techniques, which are determined at the observation planning stage. The authors of one of the articles devoted to the method point out that in most of the modern theories, statements of a dynamic nature are implicitly or directly put forward. In other words, they appeal to the justification of a certain phenomenon in the context of changes that occur with it or its connections with other phenomena. A similar conclusion can be drawn regarding the psychological patterns that are established when testing hypotheses about development, delayed or long-term effects of exposure.
Relationship with empirical observations
Hypothesis testing is the key task performed by the longitudinal method. However, despite this, conclusions about development are often made in accordance with the results of empirical observations. They are carried out within the framework of various psychological concepts using the slice method. It allows you to detect relationships between several static variables taken in a separate time period. The use of the obtained conclusions is determined by the presence of an unspoken assumption about the equivalencesamples through which comparison is carried out, as well as historical periods for various categories of subjects. This often results in an important source of confusion that needs to be given special attention.
Key concepts
To denote the commonality 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 geographical or other population, who experienced similar events in a given time period. The age variable is 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 is covered by the life of the cohort, including historical events common to its members. Formally, community is defined as follows:
Cohort=Measurement period (calendar year) – age (number of years from birth).
Explanations
The above equation illustrates the linear relationship between measurement time, cohort and age. In this case, an important source of systematic mixing is expressed for the longitudinal method. People born in the same year live in general social conditions covering a specific historical period. From this follows the following conclusion. Common to the people of the cohort will be not only the year of birth, but also their "history" - the content of the period in which they live in a certain country, in specificgeographical conditions, political, economic, cultural space. If this confusion is ignored, then the validity of the conclusions that a specialist using the longitudinal method will receive can be called into question.
Consequences
Linear dependence leads to the fact that in the course of monitoring 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 is different for the participants in the longitudinal slices and sections. This leads to a mixture 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 more mature and younger subjects may not express the line of development of the main process, but the effects of the cohort. The use of a longitudinal method with multiple consecutive measurements can help to detect results that are not set as the subject of research, but the consequences of the impact of social circumstances, as a historical stage specific to this sample.
Trying to overcome addiction
They are divided into 2 conceptual categories. The first is Mason's research. In it, the problem is supposed to be solved at the statistical level. To do this, models are formed by which collinearity (absolute mathematical dependence) between the cohort, age and time is eliminated.segment. The second group contains approaches that imply a theoretical justification for the process of excluding consideration of the impact of one indicator on the identified lines of development or their rethinking. Several methods have been developed in this direction. 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. Ideally, period and cohort effects, which have a fundamentally different explanatory status than time measures, are excluded from the analysis. They will be replaced by operationalized properties that make it possible to separate the 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 in relation to several cohorts at the same time.
Goals
The longitudinal method allows you to test "strong" causal hypotheses when performing a quantitative assessment of the dynamic properties of development. The key learning objectives are:
- Improving the accuracy of measuring the effect. It is achieved by controlling intra-individual variability. In this case, schemes of repeated observations are used, which, among others, include the longitudinal method.
- Checking hypotheses related to the direction of casual connections, assessing their strength.
- Determination of the functional form of curvesdevelopment or intra-individual trajectories.
- Analysis of interindividual differences. It is carried out with the help of casual models.
In the literature, the key difference in understanding the considered method is the lack of consensus on the issue of the minimum number of time slices.
