Stages of creating a psychological test. Belova O.V. General psychodiagnostics. Methodical instructions. Goal setting at the planning stage

Standardization(A. Anastasi) - this is "the uniformity of the procedure for conducting and evaluating the performance of the test." Thus, standardization is considered in two ways: as the development unified requirements to the procedure of the experiment and as the definition of a single criterion for evaluating the results of diagnostic tests. Standardization of the experimental procedure implies the unification of instructions, examination forms, methods for recording results, and the conditions for conducting the examination.

Stages of standardization

At the stage of developing a test, as well as any other method, a standardization procedure is carried out, which includes three stages.

The first stage of standardization psychological test is to create a uniform testing procedure. It includes the definition of the following points of the diagnostic situation:

1) testing conditions (room, lighting, etc.) external factors). It is obvious that the volume short term memory it is better to measure (for example, using the subtest of repetition of digital series in the Wechsler test) when there are no external stimuli, such as extraneous sounds, voices, etc.
2) The content of the instruction and the features of its presentation (tone of voice, pauses, speed of speech, etc.). For example, in the "10 words" test, each word must be presented after a certain time interval in seconds.
3) Availability of standard stimulus material. For example, the reliability of the results obtained significantly depends on whether the respondent is offered home-made G. Rorschach cards or standard cards with a certain color scheme and color shades.
4) Time limits for this test. For example, an adult respondent is given 20 minutes to complete the Raven test.
5) Standard form for performing this test. The use of a standard form facilitates the processing procedure.
6) Accounting for the impact of situational variables on the process and test results. Variables mean the condition of the test subject (fatigue, overvoltage, etc.), non-standard testing conditions (poor lighting, lack of ventilation, etc.), interruption of testing.
7) Consideration of the influence of the behavior of the diagnostician on the process and the result of testing. For example, the approving and encouraging behavior of the experimenter during testing can be perceived by the respondent as a hint of the "correct answer", etc.
8) Accounting for the influence of the respondent's experience in testing. Naturally, the respondent, who is not the first time undergoing the testing procedure, has overcome the feeling of uncertainty and has developed a certain attitude towards the test situation. For example, if the respondent has already completed the Raven test, then most likely you should not offer it to him a second time.

The second stage of psychological test standardization is to create a uniform assessment of test performance: standard interpretation of the results obtained and preliminary standard processing. This stage also involves comparing the obtained indicators with the norm for performing this test for a given age (for example, in intelligence tests), gender, etc. (see below).

The third stage of standardization of the psychological test consists in determining the norms for performing the test.

Norms are developed for different ages, professions, genders, etc. Here are some of the existing species norms:

school norms are developed on the basis of tests of school achievements or tests of school abilities. They are established for each school level and operate throughout the country.

Professional standards are established on the basis of tests for different professional groups (for example, mechanics of different profiles, typists, etc.).

Local norms are established and applied for narrow categories of people that differ in the presence of a common sign of age, gender, geographical area, socio-economic status, etc. For example, for the Wexler test for intelligence, the norms are limited by age limits.

National regulations are developed for representatives of a given nationality, nation, country as a whole. The need for such norms is determined by the specific culture, moral requirements and traditions of each nation.

The presence of normative data (norms) in standardized methods of psychodiagnostics is their essential characteristic.

There are requirements for the experiment:

1. Instructions should be communicated to the subjects in the same way, usually in writing; in the case of oral instructions, they are given in different groups in the same words, understandable to all, in the same manner;
2. No subject should be given any advantage over others;
3. During the experiment, additional explanations should not be given to individual subjects;
4. The experiment with different groups should be carried out at the same time of day, if possible, under similar conditions;
5. Time limits in the performance of tasks for all subjects should be the same, and so on.

Any norm usually changes over time along with the natural changes that occur in people's psychology over the years, so there is a rule according to which the norms of a test, especially an intellectual one, should be reviewed at least once every five years.

To fulfill the statistical norm, methods of mathematical statistics x are used (x - arithmetic mean, - standard deviations). X 5

A percentile is the percentage of individuals in the standardization sample.

Under the leadership of Gurevich, tests are being developed in which the reference point is an objectively set socio-psychological standard, independent of the test results. It is implemented in a set of tasks that make up the test. The test itself in its entirety is such a standard.

When developing and applying any reference point, attention should be paid to the sample of subjects. AT mathematical statistics Distinguish between the general population (population) and the sample. Any large set of people about which we are going to draw conclusions is called the general population.

Sample is a part, or subset, of an aggregate. It is not customary to conduct studies for the population. Usually a group of people is selected from it - a sample of standardization, which is actually tested, and with its help the general population is estimated. In order for the estimates to be reliable, the sample must be representative, representative of the population under consideration, that is, its probable properties must coincide or be close to the properties of the general population.

One way to ensure the representativeness of the sample is to limit the population (sex, age, profession, health, socio-economic status, etc.). Such a population is defined as specific.

1. Selection of subjects in the standardization sample
2. Definition of a population with the allocation of significant, insignificant variables (age, gender) in its structure
3. The population is divided into parts depending on the significant variables.
4. Subjects are selected randomly and in proportion to the size of each significant part of the population. The minimum sample threshold is 200 people.

Statistical norm - a criterion by which the results of diagnostic methods are compared.

Age norm - as an indicator that appears in the Bene-Simon tests (the norm is to do all the tasks).

External criterion - performance criterion.

A percentile is the percentage of individuals in a standardization sample whose primary score is lower than that primary score. Percentiles indicate the relative position of an individual in the standardization sample. The lower the percentile, the worse the position of the individual. Percentiles should not be confused with ordinary percentages, which are primary indicators and represent the percentage of tasks completed correctly, while percentile is a derived indicator indicating the proportion of total number group members.

standardization reliability validity tests

First stage development test methodology can be arbitrarily labeled as a stage formation of the information base of the study(the term of V.M. Melnikov and L.G. Yampolsky).

It includes the choice of the object and purpose of the study, the development of the concept of the test, and the definition of its scope.

Object "ect research suggests that any the property must be satisfactorily described through a system of features that define it, or external manifestations . For example, when creating a test for the study of sociability, a psychologist must find out how this characteristic manifests itself - talkativeness, the presence of many acquaintances, expressiveness, etc. In other words, the object of psychodiagnostic research is the mental phenomenon for the diagnosis of which we are developing a test.

Target is determined primarily by the range of practical problems, for the solution of which the future test methodology is being created. Developing an example with such a phenomenon as sociability, we should to decide for what purpose we are going to diagnose it: regarding age, a particular profession, success in any activity.

Region application of the test.

Traditionally, the scope of the test is considered as a characteristic of the methodology, indicating the characteristics of the contingent of subjects for whom the test is intended. For this group, norms are set, the optimal difficulty of test items, the characteristics of validity, reliability, etc. are determined. This is the scope of the test from a population perspective.

Speaking about the breadth of the population of people to whom this test is applied, one can cite as an example such methods as the so-called "culture-free" intellectual tests (which, as we will say in due time, are utopia, but for a very long time occupied the minds researchers), Luscher test. Or methods aimed at diagnosing, say, professional deformation in our penitentiary system (a much narrower circle of the population).

I distinguish the same scope of the test in terms of its content. Nap For example, such universal characterological questionnaires as Cattell's 16 PF, SMIL, as conceived by the creators, should cover the entire, or at least most of the personality structure. While the scale personal anxiety, aggressiveness affect its separate aspect. That is, the spectrum of mental phenomena that is covered by the diagnosis of this test.

The first stage ends with a description of the concept of the test, in which the main attention should be paid to the features that define the basic concept. Based on them, the interpretation of the results of the finished test is further built. The first stage, among other things, reflects the theoretical approach of the author of the test to the psychological reality under study. For example, speaking about the creation of a methodology for diagnosing personality, we are talking first of all, whether the creator of the test adheres to trait theory (as the author of the 16 PF Cattell questionnaire) or type theory (MMPI, Smishek questionnaire, ITO).

Second phase concerns the direct design of the test as a system of tasks. This stage includes the choice of test scales, determining the types of tasks, depending on the nature of the answers, compiling and formulating tasks, and placing them, grouping and numbering, compiling keys to tasks, writing instructions.

In general, this stage begins with the development stimulus material. L.F. Burlachuk defines the concept of stimulus material as follows:

objects alive and inanimate nature, artificially created objects, their images of varying degrees of accuracy, colors, sounds and symbols that serve as tasks for psychological tests.

The degree of structure of the stimulus material plays a special role. Weakly structured, ambiguous stimuli, due to the launch of projection mechanisms, carry very interesting, deep and not subject to conscious distortion material, however, it is associated with a number of difficulties in interpretation.

In verbal test methods, on the example of which we are considering the topic of test design today, verbal stimuli are used in the form of questions, statements.

At the seminar session, our speakers will tell us in more detail about approaches to the development of stimulus material.

We only add that it is determined by the direction of the methodology and is incorporated into it at the development stage (content validity).

In selection test items are most pronounced Creative skills and developer ingenuity.

Job Requirements:

Easily understood by subjects;

Be new enough for them;

Be compact, concise, not contain unnecessary information;

Do not cause additional questions from the subject;

Require a relatively short time for a response (solution);

The chance of random responses should be kept to a minimum.

In test questionnaires and most intellectual tests, tasks of a closed type are used (i.e., tasks that have a number of possible answers, among which you must choose one correct one). The simplest are tasks that involve the choice of one of two alternative solutions (the dichotomy of choice, or tasks like "yes" - "no"). The disadvantage of this kind of tasks is the high probability of random answers.

In personality questionnaires, intermediate answers of the type ("something in between", "hard to say") are sometimes used, and it is stipulated that their use should not be too frequent (such answers, due to non-discrimination, provide little information). (An example with the answer “I don’t know” in SMIL: up to 40, according to the authors, do not affect the result, but psychodiagnostics try to orient test takers to a smaller number of such answers).

Multiple-choice tasks are most typical for intelligence test batteries (Eysenck, Amthauer, Wexler); from several answers, one is selected that is correct in the subject's opinion. Often among the many (usually no more than 6-8) answer options, along with the correct ones, there are 2-3 plausible ones. Answers should be selected in such a way that each one is chosen with the same probability. The position of correct answers should change.

When diagnosing conditions, the choice of a specific point on the scale is often used as answer options. Such a scale means a gradation of the severity of a particular condition. Typically used even number gradations (for example, 4) in order to avoid concentration of answers near the middle. (for example: in USK, many diagnosticians try to omit), because its frequent use leads to averaging of the results obtained.

A special case presents the preparation of tasks for projective methods. A feature of such tasks is their unstructured, uncertainty, which allows for an almost unlimited variety of possible answers. The analysis of responses is predominantly qualitative, not quantitative, so normalization in relation to projective methods is difficult.

The necessary stage of preparing the primary form of the test is drawing up instructions.

Main requirements To her:

1) must be complete, i.e. contain, if possible, all the information for performing test tasks;

2) should not be too long. The possibilities of human memory are limited, therefore, having missed any part of the instruction, the subject may not understand it as a whole. In addition, the subject may get the impression that the test is too difficult;

3) must be unambiguous, not allow ambiguous interpretation;

4) should be designed for the weakest;

5) it is desirable to accompany the instructions with illustrative examples and samples, especially if the material is unfamiliar.

The fact that the placement of individual items in the test can affect the result of the subject, we mentioned in the topic "reliability". This applies not only to the scales of "lie". Difficult, easy and medium-difficulty tasks in the general array of the test are arranged, as a rule, in a random order. The exception is the methods, the complexity of the tasks in which is gradually increasing (among the blank methods, Raven's progressive matrices can be cited as an example).

The implementation of the above procedures is intended to create a primary form of the test, which involves:

1) stimulus material;

2) instructions for the subject on its implementation;

3) keys for processing the received data;

4) approaches to interpretation.

Third stage includes pilot studies of this form on a representative sample. It includes checking the reliability, validity and (for personality test questionnaires) the reliability of the test.

As is known, psychological test can be characterized as effective if it satisfies the following basic conditions:

1) use of the interval scale;

2) reliability;

3) validity;

4) discrimination;

5) availability of regulatory data.

Let's briefly analyze each point:

1. Using the interval scale. The interval scale is the first metric scale that allows you to introduce the concept of a measure on a set of objects. In other words, it determines the magnitude of differences between objects in the manifestation of a property. It is with the help of the interval scale that 2 objects can be compared. Classic example The interval scale is the temperature measurement scale in Celsius. The interval scale always has a scale unit, but the position of zero on it is arbitrary. Most psychological measurement theorists believe that tests measure mental properties using a scale of intervals.

2. Reliability– i.e. the accuracy of psychodiagnostic measurements, as well as the stability of the test results to the action of extraneous, random factors. Extraneous, random factors - sources of measurement error - are: the subject himself (his condition, mood, attitude towards testing, ability to concentrate, etc.); Environment, i.e. testing conditions (forms, seats, illumination and ventilation of the room, serviceability of equipment, etc.); psychodiagnostic (his mood, ability to establish psychological contact with the audience, accuracy in data processing and scoring for keys, etc.).

3. Validity- suitability; a complex characteristic of the test, including information about the area of the studied phenomena and the representativeness of the diagnostic procedure in relation to them (Burlachuk L.F.); a characteristic that tells us what the test measures and how well it does it (A. Anastasi).

4. Discriminativity- the ability of individual tasks (points) of the test to differentiate the examined relative to the "maximum" or "minimum" test result. Any response of the subject to a specific test task can be assessed on a two-point scale - "true (1 point) - incorrect (0 points)".

If all subjects give the same answer, then this means that this task does not have discrimination.

5. Availability of regulatory data - those. data obtained by comparing individual results with the statistical values of a normative sample (i.e. standardization sample). Cm. normal distribution, data normalization.

In addition, in tests of intelligence, abilities, achievements, an analysis of the difficulty of test tasks is carried out. Most often, the difficulty of the task is determined by the percentage of subjects who gave the correct answer. The easier the task, the higher this percentage, of course.

In conclusion, we can say that in our country, the technology for creating and adapting test methods was understood somewhat simplistically. The entire above process was reduced to a simple translation of one or another foreign method, at best limited to constructing a normative distribution of test indicators. Theoretical concepts of the authors of the tests were not analyzed; data on their reliability and validity were taken as true.

Then, in the 80s, the issues of adapting various foreign tests increasingly became the subject of discussion by Soviet psychologists, and later by CIS psychologists.

The requirements for the development and adaptation of tests imply a high professional culture of the psychologist, the widespread use of special techniques, including those based on modern computer technology.

Send your good work in the knowledge base is simple. Use the form below

Students, graduate students, young scientists who use the knowledge base in their studies and work will be very grateful to you.

Posted on http://www.allbest.ru/

1. Technology of test methodology adaptation: task analysis

The situation of constructing psychological tests assumes the existence of only the initial concept mental properties along with the lack of a procedure for its measurement that meets the requirements of the place, time, possibilities quantitative analysis and restrictions on other resources (A.A. Bodalev, V.V. Stolin). Despite the fact that the design of methods related to different approaches has certain specifics, a generalized algorithm of test development activities can be distinguished.

1. Theoretical substantiation of the methodology.

2. Actually designing.

2.1 Test specification (determining the number of tasks). A detailed procedure for the specification of tests on the example of questionnaires is given in the textbook by L.F. Burlachuk.

2.2 Development of tasks.

2.4 Pre-registration.

At this stage, a pilot study is carried out on a reduced sample of standardization. The main psychometric indicators are calculated. First of all, the analysis of the test items is carried out. This procedure is covered in detail by A. Anastasi. The analysis of test items (test items) is carried out according to two parameters: the difficulty of tasks (the proportion of subjects in the sample who completed the task) and the discrimination of tasks (the distinctive ability of tasks). At the stage of approbation of the methodology, a preliminary calculation of reliability and validity indicators is also carried out.

4. Calculation of psychometric indicators on the sample of standardization.

5. Standardization of the test. Calculation of test norms.

6. Final design of the methodology.

7. Revisions to methodology over time.

The situation of adaptation involves rechecking the main psychometric indicators of the methodology in new conditions. First of all, translation tests are subject to adaptation. The adaptation of foreign tests in terms of the volume of empirical and statistical work is practically not inferior to the creation of original diagnostic methods.

A.A. Bodalev and V.V. Stolin identifies the following stages of adaptation of the translation test:

1. Analysis of the internal consistency of test items.

2. Checking resistance to retesting.

3. Analysis of correlations with a relevant external criterion.

4. Verification or re-standardization of test norms.

5. In the case of adapting a multidimensional methodology, checking the reproducibility of the structure of relationships between the scales.

The situation of "intracultural transfer" of the test to a new population, which differs from the standardization sample in terms of gender and age or professional and cultural characteristics, poses special tasks for the psychologist. AT this case necessary:

1) check the validity of the methodology;

2) recheck the test norms.

2. Characteristics of the stage of processing and interpretation of the psychodiagnostic process

Stages of the psychodiagnostic process. In the psychological literature, several variants of the schemes of the diagnostic process are given, with varying degrees of completeness reflecting its real content (Burlachuk L.F., Gilbukh Yu.Z., I.V. Dubrovina, I. Shvantsara, Bodalev A.A., Stolin V.V. . and etc.).

I.V. Dubrovina, summarizing the experience of school psychologists and relying on the research of I. Shvantsar, identifies the following stages of work on psychodiagnostics:

1. The study of the requirement received by the practical psychologist in the form of a request (complaint, doubt, question) of teachers, parents, students, involving its clarification through a special conversation,

2. Formulation psychological problem, which consists in translating the request to psychological language based on the study of all information about the student.

3. Putting forward hypotheses about the causes of deviations in learning, upbringing, and development of the student.

4. The choice of a research method, which implies the availability of a practical psychologist and his possession of both strictly formalized (tests) and slightly formalized (observation, conversation, etc.) methods and research.

5. Evaluation of the results obtained, which consists in comparing them with manifestations in life situations and age specific.

6. Diagnosis, formulation of the conclusion.

7. Translation of the diagnosis into the language of the addressee (teachers, parents, student).

10. Development of ways and means of correction.

11. Implementation of psycho-correctional work.

Another scheme of the diagnostic process was proposed by L.F. Burlachuk. The diagnostic activity of a psychologist can be represented as different stages of the information processing process leading to a decision - diagnosis and prognosis. The main stages of the diagnostic process are reduced to the collection of data in accordance with the task of the study, their processing, interpretation and decision making (diagnosis and prognosis).

Stage of processing and interpretation. At this stage, the obtained data are processed and interpreted on the basis of a harmonious combination of two approaches: clinical and statistical. Clinical, close to judgment common sense and more focused on the experience and intuition of the diagnostician. Statistical provides for the accounting of objective quantitative indicators, their statistical processing. The role of subjective judgment is reduced to a minimum. The question of the effectiveness of clinical and statistical prognosis has been repeatedly discussed by psychologists and is still the subject of discussion.

Decision-making stage. N. Sandberg and L. Tyler distinguish three levels of diagnostic conclusions. At the first level, a diagnostic conclusion is made directly from the data available about the subject. (For example, it was found that the success of the tasks of the methodology allows us to conclude that there are no thinking disorders. This serves as the basis for selection). With this "selective" approach, the psychologist is not interested in why an individual subject could not complete the test items. An individual diagnosis, and even more so a prognosis, is not carried out. Diagnosis at this level is closed in a vicious circle, it returns to the clinic its own data, but only expressed in a different system of concepts. Even L. S. Vygotsky noted that this kind of diagnosis is reduced to a retelling of the initial data in other words, and is supplied with a “bright, for the most part, foreign and incomprehensible label.” Such a widespread type of diagnosis, when a psychologist can be replaced by a machine or a person specially trained to conduct testing, has been repeatedly criticized. But it should be noted that this level should be understood as a purely working, indicative, and in some cases corresponding to the tasks set (for example, the study of a significant number of individuals in order to differentiate them).

The ordering and integration of diagnostic indicators into a system is often complicated when the methods used are based on various theoretical positions. In this case, the integration of various data is carried out either at the empirical level by developing a scheme that allows for a conditional comparison of the results of individual methods (for example, by developing a five-point scale, the opposite poles of which characterize the "high" and "low" indicators of each method), or by theoretical generalization and systematization of diagnostic data based on the concepts of any one theory of personality.

The psychodiagnostic study ends with the development of a program of actions that must be carried out in connection with the results obtained, recommendations for choosing the best methods for treating the disease, rehabilitation, etc.

The results of a diagnostic study should be presented in explanatory terms, i.e., not the results obtained using specific methods using special terminology are described, but their psychological interpretation. The terms used should be "further defined" by referring to the relevant theory, such as "Eysenck introvert" or "Rorschach introvert".

Valuable in the above scheme is a fairly complete description of the content of the psychodiagnostic process from the standpoint of information processing by the diagnostician, as well as the solution of such specific problems that arise in the implementation of psychodiagnostic activities, as the identification of ways to integrate data from various psychodiagnostic methods, and the construction of types of psychodiagnostic conclusions that meet the current level of development of psychological science. designing psychological tests

The generalization of the known models of the psychodiagnostic process on the basis of the methodological provisions of the general theory of diagnosis allows us to distinguish the following stages:

1. Determination of the state of the object of psychodiagnostics at the phenomenological level. At this stage, based on the known psychological science classifications of variable parameters of activity or states of mental discomfort (classifications of objective indicators of educational or professional activity, typical complaints, problems, requests of the subjects) the recognition of the available, in this particular case, elements of the phenomenological level is carried out by their characteristics and determining their status in terms of compliance with the norm. This stage includes:

a) familiarization with the request of the subject or persons associated with him;

b) determining the compliance of this particular case with the competence of a practical psychologist;

c) collection of data on the variable parameters of activity or the state of mental discomfort, involving a survey of the subject or persons associated with him in order to clarify complaints, problems, requests and collect anamnestic data, as well as examining the client using special techniques;

d) creation of a synthetic picture of the object of psychodiagnostics at the phenomenological level;

e) an assessment of his condition and the formulation of a psychodiagnostic task.

2. Putting forward hypotheses about the psychological causes that determine the state of the elements of the phenomenological level (variable parameters of activity or mental discomfort) established at the first stage on the basis of determination schemes and psychodiagnostic tables.

3. Hypothesis testing. "Is carried out by determining the state of the object of psychodiagnostics at the level of causal grounds and involves:

a) the choice of methods of psychodiagnostics that are adequate to the hypothesis;

b) examination of the client with the help of psychodiagnostic techniques in order to determine the state of psychological variables of the level of causal grounds;

c) interpretation and assessment of the state of psychological variables from the point of view of the norm.

4. Construction general formula diagnostic conclusion. It is implemented on the basis of determining the type of psychological variables in accordance with the classifications of elements of the level of causal grounds known at the present level of development of psychological science, which determined the state of the parameters of activity or mental discomfort established at the first stage.

5. Individualization of psychological diagnosis. At this stage, the diagnosis is specified and translated into the language of the addressee,

6. Forecasting the state of the subject and formulating recommendations for the provision of psychological assistance.

7. Observation of the client after the provision of psychological assistance, followed by clarification of the diagnostic conclusion.

The validity of the outlined scheme of the psychodiagnostic process is confirmed by experimental studies performed on models of diagnostic activity that differ in content. The main stages of the proposed scheme of the psychodiagnostic process are also found in the analysis of cases from practice. psychological counseling. At the same time, this scheme should undergo additional testing in various areas of psychodiagnostic practice.

So, we can draw the following conclusions;

1. The psychodiagnostic process, characterizing the formulation of a psychological diagnosis by a practical psychologist, is a complex procedure that requires special training of a diagnostician.

2. The complexity of the psychodiagnostic procedure is expressed in the presence of a number of stages, deviations in the implementation of each of which lead to diagnostic errors, and are determined by the hierarchical structure of the object of psychodiagnostics and the ambiguity of causal relationships between its elements of different levels.

3. The psychodiagnostic process is carried out in the form of interaction between the diagnostician and the subject, mediated by psychodiagnostic means.

5. The number and content of the stages of the psychodiagnostic process depends on the psychodiagnostic tools available to the practical psychologist, as well as on the existing system of his psychological preparation, due to the degree of development of the scientific and methodological foundations of psychodiagnostic activity.

6. On present stage development of psychological science, there are several schemes of the psychodiagnostic process, revealing it from different angles and with different completeness. In this regard, the task of experimental verification of the comparative effectiveness of each of them in the psychodiagnostic work of practical psychologists is relevant.

Hosted on Allbest.ru

2. Psychological test as an objective measurement

Objectivity psychological test means that the primary indicators, their assessment and interpretation do not depend on the behavior and subjective judgments of the experimenter. Primary indicators are those obtained after processing the data on the respondent's performance of test tasks. Primary indicators are expressed in the so-called "raw points".

The development of tasks, the procedure for conducting, processing the test results is carried out according to certain standard rules.

The objectivity of a psychological test can be achieved if the following conditions are met:

uniformity of the test procedure to obtain results comparable to the norm (see below);

uniformity of assessment of test performance;

determination of the test performance norm for comparing with them the indicators obtained as a result of processing the test data (see here "the third stage of standardization").

These three conditions are called standardization stages psychological test.

Stages of standardization

At the stage of developing a test, as well as any other method, a standardization procedure is carried out, which includes three stages.

First stage standardization of the psychological test is to create a uniform testing procedure. It includes the definition of the following points of the diagnostic situation:

testing conditions (room, lighting, and other external factors). Obviously, short-term memory capacity is better measured (for example, using the Wechsler digit repetition subtest) when there are no external stimuli, such as extraneous sounds, voices, etc.

The test creation process scientific justification, revisions and improvements can be broken down into a number of steps.

1. Determining the purpose of testing, choosing the type of test and the approach to its creation.

3.Determination of the structure of the test and the strategy for arranging tasks.

4. Development of test specification, a priori choice of test length and execution time.

5. Creation of pretest tasks.

6. Selection of tasks in the test and their ranking according to the chosen presentation strategy based on a priori author's estimates of the difficulty of tasks.

7. Examination of the content of pre-test tasks and tests.

8. Examination of the form of pre-test tasks.

9. Reworking the content and form of assignments based on the results of the examination.

10. Development of approbation testing methodology.

11. Development of instructions for students and for teachers testing the test.

12. Carrying out approbation testing.

13. Collection of empirical results.

14. Statistical processing of test results.

15. Interpretation of processing results in order to improve the quality of the test. Verification of test performance against evidence-based quality criteria.

16. Correction of the content and form of tasks based on the data of the previous stage. Cleaning up the test and adding new items to optimize the range of values for the difficulty parameter and improve the backbone properties of the test items. Optimization of test length and execution time based on a posteriori estimates of test characteristics. Optimization of the order of assignments in the test.

17. Repeating the testing phase to perform the next steps to improve the quality of the test.

18. Interpretation of processing data, establishment of test norms and creation of a scale for evaluating the results of the subjects.

A kind of cycle arises, since after cleaning the test, the developer has to return to the stage of collecting empirical data, and, as a rule, not once, but two, three or more. In a certain sense, this cycle is endless, but not because all tasks are bad and the developer does not have enough experience in creating a test. It's just that the process of designing a test is rather difficult, since a comprehensive approach is needed to assess the quality of the test and the characteristics of test items, to check their system-forming properties.

In addition, we can assume that the problem of optimal selection of the composition of the test does not have a single solution, since not everything here is determined by the quality of the test material, but much still depends on the level of preparation of a group of students. Tasks that work well for one sample of students may be completely useless for another, as they will be either too easy or too difficult and not a single student in the group will perform them correctly.

The success of creating a test largely depends on the high quality of the initial test material, which is ensured by the correct selection of the content being checked and the ability of the developer to correctly display it in the test tasks. The stage of processing empirical test results is extremely important, for which special software tools are required for professional test development.

Of course, not always in the process of creating tests it is necessary to achieve professional level quality, especially if we ignore the goals of admission of applicants and certification of graduates. In the daily activities of the teacher, tests of different, sometimes low quality, task-oriented tests are needed. current control. The fulfillment of the last task is quite within the power of an individual teacher or a group of teachers. However, even in this case, you can independently make a number of calculations based on the minimum mathematical apparatus and allowing you to significantly move from the pre-test tasks to the real test.

findings

1. The theory of pedagogical measurements is interdisciplinary in nature, combining the achievements of pedagogy and psychology with the achievements of mathematics, statistics and measurement theory.

2. The development of the conceptual apparatus of the theory of pedagogical measurements is important for improving the quality of tests.

3. When formulating the conceptual apparatus, it is necessary to classify the types of tests in order to bring the introduced definitions into line with various types of tests and the purposes of their creation. .

4. Pedagogical tests can be used for input, current and final control when evaluating the results of the implemented educational process.

5. Pedagogical tests are developed within the framework of two approaches that allow different interpretations of the results of their performance.

6. The observed results of the test are obtained as a result of the interaction of many subjects with many test items.

7. Professionally designed tests provide an indication of the true scores of students, determined using special methods on the observed test results.

8. The process of constructing a test includes a number of stages, the implementation of which is mandatory. The exclusion of any stage leads to the inevitable loss of test quality.

QUESTIONS AND TASKS

1.What are the functions of entrance testing? Does it make sense to develop entrance tests at school?

2. What are the goals of developing formative tests? Are there differences between formative tests and traditional monitoring tools?

3. What is the purpose of the final test?

4. What approach, in your opinion, should be used to develop tests for final exams at school?

5. Does your school evaluate the performance of teachers correctly?

6. What process is called test standardization?

7. List the factors that affect the stability of the test norms.

8. What tests, in your opinion, should be developed in the first place to improve the effectiveness of the educational process at school?

9. Formulate the definitions of the pre-test task, test task, pedagogical test. Compare your answer with the content of the relevant sections of the manual.

10. What are the advantages of the pre-test task in comparison with traditional control tasks?

11.What are General requirements for pre-tests? Compare your answer with the list of general requirements provided in the manual.

12. Describe the factors that reduce the accuracy of test measurements.

13. Is it possible to get an idea of the true scores of students using traditional controls?

14. Three students answered 6 test tasks, ranked by increasing difficulty. As a result of the answers, the following profiles were obtained:

first: 111000; second: 101010; third: 000111.

Who, in your opinion, learned the content of the tested course better? How many errors are in the response profile of each of the three students? Which of the three students will have a higher true score? Is it right to pose the last question in relation to the results of the third student?

15. List the main stages of test development.

Goal setting at the planning stage

When creating a test, the attention of the developer is primarily attracted to the selection of content, which can be defined as the optimal display of content. academic discipline in the test system. The requirement of optimality implies the use of a certain selection methodology, including the issues of goal setting, planning and assessing the quality of the test content.

The goal-setting stage is the most difficult and at the same time the most important: the quality of the test content primarily depends on the results of its implementation. In the process of goal-setting, the teacher needs to decide what results of students he wants to evaluate with the help of the test. The answer would seem simple, at least it seems so to those who repeatedly test students' knowledge in the classroom by traditional means. However, in reality, this apparent simplicity often turns into a low quality of control results, when students of different backgrounds receive the same marks, or the teacher comes to the wrong conclusion about the achievement of learning goals, while students did not receive the best results. important knowledge or did not learn how to use them.

The grounds for errors in the conclusions of the teacher are not always associated with the technological shortcomings of traditional means of control. Sometimes they are due to the teacher's shortcomings at the goal-setting stage, when the center of gravity of the test is shifted to secondary learning goals, and sometimes the goal-setting stage is completely absent, since some teachers are confident in infallibility; their experience and intuition, especially on the condition of many years of work at the school. However, no even very perfect control methods and no experience will give grounds for reliable conclusions about the achievement of learning goals as long as there is no confidence in the correct setting of control goals and in their correct, unbiased display in the content of the test.

Unlike the selection of the content of traditional means of control, which is mainly done intuitively based on the practical experience of the teacher, the selection of the content of the test has a clear target orientation, and this, subject to the correct setting of goals, is a serious bid for its high quality. Figuratively speaking, when creating a test in the mind of the developer, the content of the control is refracted through the prism of the set measurement goals, and if they are formulated correctly, then there is a greater confidence that the test will take place.

True, the correct setting of the goals for creating a test is rather difficult in itself, and besides, the situation is complicated by a number of circumstances. One side, various schools in modern pedagogy use when setting goals. different systems concepts and ideas, often poorly combined with each other. On the other hand, the use of different words and expressions of natural language as terms gives rise to a fair amount of confusion, often the context of a number of pedagogical approaches is insufficient for the implementation of the term-forming function.

At the same time, for a number of years, there was simply no particular need for the formation of unified approaches to the formulation of goals. The predominant focus on traditional means of control that existed in our country for decades eliminated the need for this process, since the vagueness and uncertainty of existing educational goals did not come into conflict with the requirements of practice, far from the problems of creating objective measuring instruments. And finally, the situation was often complicated by the researchers themselves, who, figuratively speaking, prophesied, but did not hear each other. In general, this has led to the fact that so far not developed general rules the choice of grounds for the classification of goals, no principles shared by all were found for quantifying the degree of their achievement.

Thus, the ideas that have developed to date about the most common goals of control do not allow one to proceed directly to the development of measuring instruments. The formulation of educational goals is characterized by excessive generality, vagueness, diversity and uncertainty. Therefore, in order to create measuring instruments, first of all, preliminary operationalization of goals is necessary.

The process of operationalization consists in giving the content and form of presentation of the goals characteristics that make it possible to display the goals in the content of standardized measuring instruments. The idea of operationalization is very close to certain provisions of the work of M.V. Klarin, where instead of the word "operationalization" "a different, rather successful term" concretization" is used. Of course, the point is not in the term, but in the essence of the proposed process.

Stages of creating a psychological test. Belova O.V. General psychodiagnostics. Methodical instructions. Goal setting at the planning stage

Send your good work in the knowledge base is simple. Use the form below

Similar Documents

2. Psychological test as an objective measurement