Measuring procedures and tools - Qualitative and quantitative...

Measuring procedures and tools

Like any scientific ritual, the research procedure is subordinated to a technical scheme borrowed from the natural sciences. This scheme, starting to act, itself becomes an object and ceases to depend on the subjectivism of the researcher. The experimental method develops, at least in three directions. The first way is based on the reduction of the measured quality to the operational variable, the second is the preservation of a known distance between the operational constructs and the true parameters of the measured object, the third - "is characterized by an amazing and bizarre occurrence of" true " quality from the factorization of the variables

The ideas outlined in the work of G. Fechner Elements of psychophysics (1860), had a huge impact on psychological research in general and formed the basis of one of the directions of measurement procedures.

In parallel with the Fechner approach to the study of the psyche of the individual, the mathematical method developed. Differential psychology from the very beginning developed as a quantitative discipline that studied not the causal (causal) but stochastic (probabilistic) regularity development of the psyche. The statistical approach was put forward as a means of transforming psychology into an exact science. In this connection, J. Cattell wrote: "Psychology can not become solid and accurate, like physical sciences, unless it is based on experiment and measurement."

Measurement practice dates back to the origins of science, but the logical grounds for the measurement were not studied until the end of the 19th and beginning of the 20th century, when Helmholtz presented the basic ideas of the representation theory of measurement, and O. Holder developed the axiomatics of measuring extensive quantities.

Mathematics has begun to be actively introduced into psychological research and especially to testology.

The history of nonparametric methods began with the work of J. Arbetnoth in 1710, when he used the criterion of signs to test the hypothesis about the equality of the probabilities of the birth of boys and girls. In the XIX century. G. Fechner and F. Galton began to apply ranks and coefficients of rank correlation. The work of C. Spearman (1904) on rank methods attracted the attention of the psychological community, and AN Kolmogorov (1933), N. V. Smirnov (1935), F. Wilcoxon (1945), S. Siegel (1956) and other researchers have created nonparametric statistics as an independent branch of mathematical statistics.

The founder of parametric measurement in psychology is F. Galton. His views were based on the works of A. Quetelet - one of the creators of modern statistics, where the author emphasized the correspondence of people's behavior to certain mathematical laws. In particular, he showed the possibility of a probabilistic forecast of human behavior, based on average indicators. The norm of behavior was based on the average value of the behavior of the population in the form of a normal distribution curve (the Gauss-Laplace distribution law).

In 1869, F. Halgon's book "Hereditary Genius," which set forth a number of original thoughts on the use of statistical methods in the study of human abilities, was published. F. Galton, together with A. Quetelet, determined the problems of deviations from the normal curve and concluded that it is possible to estimate the probability of these deviations from the "mean pattern" abilities. He argued that the cause of such deviations are factors of heredity. His conviction passed through many of his articles and works, united under the common title "Studies on Human Abilities and Their Development" (1883). The proclamation of the regularity of the connection between the heredity of the individual and his abilities led F. Galton to the bosom of the "eugenics" and firmly linked his name to racist attitudes in the possibility of using psychodiagnostics for political purposes. Currently, the criticism of his ideas continues, often extending to a quantitative approach in general.

One of the invaluable merits of F. Galton was the development of methodological tools of psychology. The most promising method was the method of calculating the correlation coefficient between empirical variables, which was later improved by the English mathematician K. Pearson and served as a basis for the development of factor analysis.

One of the founders of objective, empirically oriented social psychology is F. Allport. His main contribution to the problem of social measurement methods was that he tried to connect the actual manifestations of behavior with the personal constructs of the subjects.

Thus, by the beginning of the 1930s. the authors of the social dimension theories tried to find the transition "from the eclectic practice of collecting empirical data [...] to direct standard ways of measuring" subjective indicators "& quot ;. However, the undeveloped sampling and measurement problems made these attempts vulnerable to criticism. These problems consisted of searching for methods for quantifying psychological information, solving problems of ensuring the external validity of the experiment and measurement, the problems of representativeness of the experimental sample, the compilation of the measurement scales themselves (the work of E. Bogardus, F. Allport, L. Thurstone, R. Likert, etc.) The criticism of opponents of the social dimension. L. Thurstoun perfected the paired comparisons method and in 1929, together with E.Chave, created a simplified procedure of "equal intervals", based on the subjective metric of the assumption of the "perceptual equality of intervals" . The method became classical and had a huge impact on the development of measurement in psychology.

In 1932 R. Likert proposed an alternative to the Thurston scale, excluding the use of expert estimates. In his summation method of ranks The procedure for calculating discrimination items for marginal groups allocated by the total score. The Likert scale was more convenient when used in mass surveys, in contrast to the Thurston, more reliable for a small number of points.

With. In 1946 Stevens published a method for direct scaling of psychological variables. He argued that there is an isomorphism between the properties of numerical series and the empirical operations that we can produce with objects. Initially, S. Stevens proposed four types of correlation of the numerical system with empirical data, which determined four corresponding scales (or levels) of measurement: the scale of names (nominal), order (ordinal), intervals (interval) and relations. Each scale allowed certain operations and mathematical transformations. Nominal scale - equality-inequality; Ordinal - equality-inequality and more-less, i.e. ranging; interval - equality-inequality, more-less and equality-inequality intervals, i.e. introduction of measurement units of signs; relations - equality-inequality, more-less and equality-inequality of intervals and relations, i.e. the implementation of all arithmetic operations.

Developing the ideas of S. Stevens, K. Kumbus in 1952 carried out an equivalent in the mathematical sense of the approach to the differentiation of scales by distinguishing the nature of arithmetic operations. E. Nigel proposed a strict formulation of the nominal scale and the scale of relations by means of axioms.

A significant contribution to the development of the psychological dimension introduced P. Lazarsfeld. He was especially interested in the problems of validity of subjective measurement; generality (identity) of psychodiagnostic indicators, providing an adequate understanding of the meaning of test questions for subjects (1935); application of analytical typologies for the classification of data "case analysis" with the subsequent creation of new variables; use of four-cell conjugacy tables to evaluate categorical variables.

The key place in the psychological dimension is the scaling model, which means the scoring method, determining the level of the measurement received (type of scale) and choosing ways to assess the functional unity of the measurement tool .

During the implementation of direct and indirect scaling models, various research situations and formal models were identified. For example, in the process of direct ordinal scaling, the coefficient of M. Kendell's concordance was used, which makes it possible to evaluate the consistency of ranks (1-the situation of complete agreement, 0-the situation of complete disagreement). S. Stevens stressed that for measuring psychological phenomena, direct scaling based on a direct evaluation of the magnitude of the stimulus by a subject is justified and gives an agreed-upon procedure. The method of successive intervals, proposed in 1937 by M. Safffir, was based on the law of categorical judgment of W. Thorgerson. The essence of indirect scaling of this type was reduced to fulfilling the task of assigning an object to the most suitable category on the continuum of preference.

Thus, the problem of formalizing research methods by the early 1960s. entered the phase of transforming empirical humanitarian knowledge into a "precise" form. scientific discipline, which allowed from the description of psychological phenomena to move to their explanation and practical application.

Psychology, all its experimental and qualitative methods, both in the past and today, being, as Eisenk called, "abnormal science", constantly encounters methodological deadlocks, moral and ethical and organizational-methodological obstacles . And "hardly anyone today decides to assert that little has changed in the field of measuring (estimating) individual differences since the days of Galton and Cattell."

In psychological science, unlike the natural sciences, where experiment and measurement are firmly on their feet and are the basis of research methodology, periods of increasing interest in the quantitative approach and activating empirical research are often replaced by a pessimistic attitude toward them. Researchers observe the manifestation of two successive forms of psychological thought, the so-called humanistic-descriptive and empirically-explanatory & quot ;. In mutual struggle they proceed from pragmatism and are strongly influenced by the prevailing social situation at the given time.

A compromise position that solves the problem of the above-mentioned "oscillations" is to find a "wise combination of qualitative and quantitative approaches". Back in the 1930s. Kurt Levin, an active supporter of experimental methods, emphasized the connection between the research problem and theoretical expediency, the latter being given the leading role. In his opinion, the experimental method should follow from the theory. Later, in the 1940s. This point of view was supported by R. Lippit, M. Raik, I. Chain, S. Cook et al.

Note that at that time, the subject area of ​​the experimental method was the identification of psychological phenomena, the process of experimentation (choice of indicators and evaluation of the organization of research), as well as the problems of statistical analysis of the results of experiments, their planning ("design") and the development of experimental schemes . There was a need for active interaction between philosophers, sociologists, historians of science and psychologists. The attention of researchers was attracted to "experiments in the field" (quasi-experiments), as well as the theory and practice of psychological measurement (validity, error, reliability, etc.).

In the problem of constructing an experimental design, an important place was occupied by the analysis of the possibility of determining the reliability of cause-effect relationships. The question was raised about the proof of the hypothesis through a quantitative assessment. Another important and controversial element in the development of experimental schemes was the procedure of "group equalization before experiment". In 1923, W. E. McCall in the work "How to conduct experiments in pedagogy" first described the groups, equalized by chance. RA Fisher in 1925 pointed to the need to "equalize groups before experiment by randomization". Randomization has become, within confidence limits, the role of primary testing. There are data that, in order to ensure the equivalence of the sample, the Latin square

was used as early as 1916.

In designing the research, control group had a significant place to increase their internal validity. It was her presence that distinguished the psychological experiment from the natural-science one. On the need to add a pilot sample to the experimental sample as early as the beginning of the 20th century. pissing X. Pierson. The "control group", according to E. Boring, was introduced into a psychological experiment in 1907. Already in 1912, in psychological studies, "control groups were used as a matter of course" and their presence in the structure of the experiment was considered an axiom. The control sample constantly improved and developed, as shown by the works of R. Rosenthal, M. Rosenberg, V. McGaire et al.

An important way to ensure the validity of the experiment was tricks "disguises & Unmasking psychological research. They were actively used by S. Ah, V. McGuire, H. Kelmen, M. Argyle, M. Rokic, D. Stollack, E. Aronson, Karlsmith et al.

P. Wolfe proposed to work out a decision about the validity of the procedure of the "cross-poll" of the experiment participants. X. Volcott for these purposes actively used the method of observation. In general, many authors - supporters of the primacy of a qualitative tradition in psychology recommended the development of experimental plans "on the basis of common sense and non-mathematical considerations."

The main lever of validating psychological research has become the ways of selecting experimental schemes (experimental design). So, back in 1782. L. Euler first used the "Latin square as a schema (plan) of scientific research. In psychology in 1916 E. Thorndike, W. McCall and J. Chapman used it in the form of 5 x 5 and 2 x 2, to provide control in the formation of equivalent groups. Later (1949) R. Solomon proposed the scheme of the "Latin square" for the four groups, which formed the basis for the development of experiments on plan 5 (according to J. Campbell). A distinctive feature of this experimental design was a high control of factors of external validity.

Design temporary series was typical of many classical natural science studies of the XIX century. In social studies of that time, it was used extremely rarely. In physiology, Pavlov's works were based on this scheme. Only in the 1920s. English scientists E. Farmer, R. Brooks and E. Chambers (1923) used this plan in studies of industrial fatigue. In the 1950s. in social psychology this scheme was actively used by A. Maxwell and B. Underwood.

The scheme "with series of equivalent impacts" was first used in A. Josga's studies in 1895-1896. and H. Moore in 1921.

E. Thorndike and R. Woodworth in 1901 first constructed and applied an experimental plan with an non-equivalent control group.

The term & amp; balanced plans experiments B. Underwood used in his works (1949), although W. McCall, calling them "rotary experiments", applied similar schemes in psychological research even earlier ( 1923). In 1957, V. Cochran and D. Cox (" cross-plans"), used the balanced plans, and O. Camphor called them switching plans

Designs with preliminary and final testing on different samples ", , according to J. Campbell, were the best among the experimental schemes. They were actively used by S. Stahr and H. Hughes in the 1940s, but their biggest drawback was the uncontrollability of the "background".

Since 1949 R. Solomon, K. Honand and F. Sheffield have applied experiments on "patchwork" plan (recurrent institutional cycle). It was first conceptually worked out in the late 1950s. It was used, in particular, by J. Campbell and T. McCormack in the study of the influence of the command flight training of US Air Force cadets on the formation of social attitudes and leadership functions in small groups.

Under the influence of JW Tukey in the 1950s. The models of variance analysis for a finite number of factor levels in the experiment were developed. In 1956, H. Scheffe published a historical review of the development of factorial plans. In the same year, J. Stanley applied a three-factor experimental plan. In 1959, J. Ferguson presented a simple explanation of the model of cases of one or several factors with fixed or random levels.

The term " experiment expost facto" was first introduced by F. Chapin in 1937. This scheme he actively used during his studies at school, trying to simulate an experiment under the plan statistical group comparison . As an independent variable in the experiment, the fact of the high school graduation was adopted, and the dependent variable was the success in life and the social adaptation of graduates after 10 years.

Thus, a psychological experiment at the turn of the XIX-XX centuries. has acquired an individual status of the main method of psychological research. Unlike the natural-scientific forms, he included stochastic methods of evidence for the causal model of the experiment. In its structure, a special place was occupied by means of ensuring internal and external validity of psychological research.

Under the influence of active introduction in the psychology of the experiment, the status of the psychological science itself has changed. "For several decades," SL Rubinshtein wrote in 1946, "the actual experimental material psychology has at its disposal has grown considerably; the methods by which it works have become more diverse and more precise; the appearance of science has significantly changed. The introduction into psychology of the experiment not only armed her with this new, very powerful special method of scientific research, but in general she raised the question of the method of psychological research in general, setting new requirements and criteria for the scientific character of all types of experimental research in psychology. That is why the introduction of the experimental method into psychology played such a big, perhaps even decisive, role in the design of psychology as an independent science. "

However, we should not forget that along with quantitative methods based on experiment, the descriptive ( qualitative) method also develops (see Chapter 2 of this textbook). Unlike experiment, his experience and result, according to his adherents, do not require any evidence. He simply describes the phenomenon, trying to understand it, to comprehend and accept it. "What has developed in psychology," wrote R. Bandler, "are different systems of religious beliefs with very powerful evangelists who work from all these different orientations."

thematic pictures

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)