Semantic differential in the study of the image of money. The use of semantic differential in marketing research
PROJECTIVE METHODS
Psychosemantic method as a method of measurement and as a method of analysis. Semantic differential of Ch. Osgood. The method of unfinished sentences as an approach to the study of social expectations and social stereotypes. The test of twenty "I" as a technique for studying the self-identification of a person.
The index analysis and ranking procedure will demonstrate that it is sometimes difficult to attribute the "method" of sociological research to either measurement methods or analysis methods. It all depends on the context, the research situation and from the goal for which method is applied. The absence of a coherent conceptual apparatus in sociology is an internal specificity and property of our science. Therefore, some methods are multivalued:
1) they act as measurement techniques,
2) as methods of analysis.
Each such procedure involves a specific technique for collecting information and a specific mathematical processing of empirical data. Therefore, the concept diagnostic procedure in contrast to the concept of method and technique is more acceptable. Unfortunately, this concept is not widely used in the sociological literature.
These include psychological procedures, borrowed by sociologists from psychology. These procedures can be called tests .
Some tests seem to measure personal characteristics, others - group characteristics. There are a lot of tests. Consider the procedures related to the so-called projective methods.
When studying social reality, the researcher asks the question of how this very reality is perceived by specific people, refracted in their minds, turns into certain social norms, images. To penetrate into inner world It is impossible for a person to use predetermined schemes, unambiguously interpreted concepts and categories. Accordingly, rigidly formalized, structured methods of collecting information do not work in this case. There is a need to involve concepts and categories that people themselves use to streamline their own daily life experience.
These methods are based, as it were, on the projection of the subjective properties of the individual onto the sociologist's screen. The sociologist offers the respondent stimuli (signs, texts, pictures, situations) and determines hidden, unconscious thought processes, needs, images, etc. by reactions.
Let's remember the logical square. To some extent, it is a projective technique. Offering the respondent two questions, we revealed, as it were, his subjective inclinations, inclinations in situations that were unrealistic for him. According to these personal values determined the meaning: the degree of satisfaction with studies as a force of motivation.
As an example, consider three projective methods:
1) Ch. Osgood semantic differential method(Ch. Osgood), is used in sociology to solve a wide range of problems and is interpreted both as a method of measurement and as a method of analysis.
2) unfinished sentence method¾ is used in sociology as an approach to the study of social expectations and social norms.
3) the test of twenty "I" ¾ is used to study the self-identification of a person. In the last two cases we are talking about working with textual information, which also determined the choice of these methods.
Ch. Osgood Semantic Differential (SDO)
This method was developed in the mid-1950s by C. Osgood to study the emotional attitude of people to certain concepts in order to determine their meaning. It is well described in the literature (for example, a review can be found in ). The SDO method is as follows. Respondents are asked to express their attitude towards some object (concept, image) using a combination of bipolar scales, mostly seven-point scales, for example, such as:
The extreme positions on the scales are described by verbal antonyms. The set of scales forms the initial space scales. The number of gradations on the scale may be less than seven. Extreme positions can also be non-verbal. For example, C. Osgood used signs "black circle ¾ white circle", "arrow up ¾ arrow down" etc. when studying representatives of different linguistic cultures (Indians, Mexicans, Japanese and Americans) in their attitude to different concepts.
Why were the scales listed above selected? Different experiments with different sets of bipolar scales gave the same result. The whole set of scales, as it were, fell into three main groups, into three factors named as Strength, Activity, Attitude .
This phenomenon was discovered by C. Osgood and named synesthesia. For our purposes, what matters is how it manifests itself. Factor¾ a group of scales according to which the assessments of the object are similar. Of the above seven scales, the “strength” factor includes scales 1 and 2 (weak ¾ strong, male ¾ female), the “activity” factor of the 3¾4 scale (active ¾ passive, slow ¾ fast), the “attitude” factor 5¾7 (normal ¾ unusual, false ¾ true, good ¾ bad). Similarly, one can argue for the case of the rest of the 20 classical scales introduced by C. Osgood. In addition to the seven listed pairs, these twenty include the following pairs:
cruel ¾ kind,
curve ¾ straight,
loose ¾ punctual,
tasty ¾ tasteless,
unsuccessful ¾ successful,
hard ¾ soft
stupid ¾ smart,
new ¾ old,
unimportant ¾ important,
sharp ¾ rounded,
cold-blooded ¾ enthusiastic,
colorless ¾ colorful,
uncommon ¾ common;
beautiful ¾ ugly.
These scales were selected after a huge number of experiments with different pairs of adjectives through the use of mathematical methods factor analysis (you don't need to think about that yet). It is for this set of scales that their three-factor structure was proved. In other words, we, as respondents, evaluate any object (concept, image) presented to us emotionally, mainly by three factors or in three-dimensional space. This space is called semantic space . For this reason, the term "semantic" is used in the name of the method. The images of objects in the semantic space occupy a well-defined specific position. Analyzing the location of objects, one can draw conclusions about the proximity of the images. For an analogy, we recall school geometry and the concepts of “two-dimensional space” (it is simply called a plane) and “three-dimensional space”.
The idea of the SDO method was further developed. In real research, the sociologist has three possibilities when working with this approach to the study of social reality, or three strategies for the formation of a set of initial scales:
a) use well-known, as it were, classical scales with slight adjustments;
b) use the results of the work of other researchers;
c) try to form your own semantic space of factors.
In the first case correction is required to eliminate pairs that are incorrect when evaluating specific objects. Let us give an example of one problem, in the solution of which the SDO method was used. The task is to conduct a typological analysis of the perception of images of 10 popular politicians, i.e., to identify different groups of politicians. At the same time, the image of politicians assigned to the same group is the same. The objects of the study were teachers of the sociological faculty of one of the universities. The study was purely methodical character, so there was no problem of representativeness and the sample was small. The classical set of scales was used, with the exception of pairs like bad ¾ good, smart ¾ stupid, etc. Such pairs were excluded due to their excessive specificity and incorrectness for the case of our problem.
Second strategy, namely the use of the results of other researchers, is possible in the following situation. Suppose that we are talking about mass surveys and the sociologist does not have the opportunity to experiment with a large number of scales and form his own semantic space. Then, in order to select the scales for his research, he proceeds as follows. For example, in the task of studying the images of politicians, we chose the same number of scales for each of the three factors. And about what factor and with what weight the scale belongs, we learned from the literature.
Third strategy¾ the formation of one's own semantic space for evaluating objects arises in deep analytical studies, when LMS is the main approach to the study of social reality. Then, initially, the sociologist forms a set of scales that are concrete, and not associative in nature (like 20 classical ones). In this case, he is obliged to test the hypothesis about the existence of factors and find out what they are and how many of them. This is necessary for the transition to the analysis of images in the space of these factors. In each task, for a set of objects specified by the researcher, there can be its own factor space with a different number of factors.
Regardless of the chosen strategy for using the LMS method, the analysis of object images is carried out in the following way. Let's talk about it on the example of studying the images of political leaders. Having received the initial data ¾ of the assessment of each teacher for each of the ten political leaders, ¾ calculated average grade for each of the three factors for each object (leader).
The rating for a factor for an object is equal to the sum of ratings for all scales included in this factor and for all respondents, divided by a value equal to the product of the number of scales and the number of respondents.
In this case, instead of writing a simple formula for calculating the average grade, we verbally described its content. As you can see, this is inconvenient. Therefore, the sociologist needs mathematics, because it gives comfortable description language.
After calculating the average score for all factors (strength, activity, attitude) separately, the transition to the calculation of proximity between political leaders is carried out. This can be done in three-dimensional space, or it can be done in two-dimensional space (choosing two factors that are important for you from three).
The second case is easier. Moreover, from school geometry you remember how the distance between points (political leaders) is calculated on a plane (two selected factors form it). Let's use the formula. Denote by D (A, B) distance between two political leaders A And IN in the space of factors, through d i (A,B)¾difference in average grades for i-mu, factor. The number of factors is equal to k. Then about the proximity of the images of political leaders A And IN can be judged by the meaning of the so-called differential.
Having carried out calculations for all pairs, and there will be 45 of them for ten political leaders, we obtain the so-called proximity matrix or a matrix of the “object ¾ object” type. Recall Thurstone's method of paired comparisons. There were also proximity matrices, only of a different nature.
If anyone was interested in our study, it should be noted that four types of images of political leaders were revealed. The study was conducted in March 1996. The image of President B. Yeltsin was different from all other images. The situation with V. Zhirinovsky is similar. The images of V. Chernomyrdin, A. Lebed, G. Zyuganov were close. All other political leaders formed the fourth group. The interpretation of this result is not important to us. This problem is of interest to us only as an example illustrating the application of the LMS method in mass surveys.
2. Method of unfinished sentences (MNP)
An example of a study that was also methodological in nature It was associated with the problems of students' subjective perception of the image "cultural person". The task was to determine the meaning that people put into the concept they use ¾ "cultured person". In everyday speech, you can often find this expression. What are people putting into it? What kind of person do they call cultured and what are the criteria for culture? What is the image of a "cultural person" in their minds? Are there different types (kinds) of perception of this image? If they exist, what are they? What is the semantic space of image perception and what is its dimension?
It is quite clear that attempts to find a clearly formulated understanding of the image in the respondent will be futile, since in the minds of people this image has blurry, amorphous outlines.
Application MNP gives the greatest effect just in those studies where there is a need to identify a person's subjective perception of social reality, personal meanings individuals, stereotypes, images, standards, value orientations of people, etc.
Respondents are offered a set of unfinished sentences with a request to write their completion. The verbally expressed reactions of the respondents to the first part of the sentences constitute the base on the basis of which it is possible to identify the main characteristics of the studied image. We can say that these reactions carry information about the norms, values, stereotypes, standards, images that exist in society and assimilated by the individual. Their reconstruction is carried out on the basis of the information collected during the survey. The following are just some of the unfinished sentences we used in this study.
1) Distinguishes a cultured person from other people ...
2) A cultured person must ...
3) Usually cultured people ...
4) The closest thing to the concept of "cultural person" is the concept of ...
5) All cultural people are united ...
6) I call a cultured person ...
7) The most important thing for a cultured person ...
8) A cultured person never...
9) A cultured person is always...
10) A cultured person has ...
11) The opposite of a cultured person...
12) I cannot call a cultured person that ...
In these sentences, the respondents retain the opportunity to choose the criteria themselves and determine the semantic basis of the answer; the reactions of the respondents are not limited to predetermined options. In general, one can count on the fact that the developed proposals make it possible to obtain the reactions of the respondent, minimally distorted by the influence of the researcher. The subject is forced to speak in his own words, as a result of which, at the end, he uses the categories that he uses in Everyday life to organize your life experience.
It was assumed that when composing sentence endings, the respondents had to use different justifications. In the broad sense of the word justification ¾ is a moral prescription, stereotyped patterns of goals, motives, etc. set by culture. In a narrow sense, it is convenient to use the concept "elementary justifications" to designate semantic particles, which are obtained, as it were, by separation text (sentence endings) into separate indivisible parts.
At the first stage of the analysis, the sentence completion texts for all respondents are divided into elementary justifications. After that, justifications similar in meaning are grouped, as a result of which separate elements , which express one or another characteristic of the image.
For example, justifications like: « Distinguishes a cultured person from other people ... ": "... the style of speech", "... the ability to clearly express one's thoughts"; "Usually cultured people...":“... they know how to explain their point of view without resorting to shouting”, “... they don’t shout at others, they don’t swear”¾ clearly have a common semantic focus and refer to the same element.
Accordingly, they differ from type justifications: “What distinguishes a cultured person from other people is ...”: “... the presence of principles”, “... the will”, “... a high moral level”; "Usually cultured people...": "... do not hesitate in making moral decisions." These justifications also constitute a separate element,
The first element marks the manner of communication of a “cultured person”. It can conditionally be called "Speech and Thought" . The second ¾ moral principles and the inner world of a "cultured person", and it can be called conditionally "Inner World" . Some elements are generalized at an even higher level, thereby forming concepts more high level abstractness. The singling out of elementary substantiations, and then the elements, is something else than logical formalization when analyzing texts. The next stage of the analysis consists in comparing the image of a cultured person for different respondents. For example, based on the frequency of occurrence elementary justifications And elements . But here it is impossible to do without quantitative calculations, without mathematical formalization . This requires a complex coding of information beforehand.
Due to the fact that so far we are mainly interested in measurement problems, we present a fragment related to this problem. As noted, the sentences “The closest thing to the concept of a “cultural person” is the concept ...” and “The opposite of a cultured person ...” allow us to establish concepts similar to the image under study and opposite to it. Most of the answers are the same images as the "cultured person". So, the respondents named the closest to this concept: “ intelligent person¾ 37%, "well-mannered person" ¾ 16%, "polite" ¾ 11% and "educated" ¾ 9%. Opposite concepts: "boor" ¾ 28%, "uncivilized person" ¾ 13%, ignoramus ¾ 8%. By themselves, these data are not very informative. Nevertheless, it is possible to raise the question of constructing a scale of "cultured person" - "uncivilized person". You can also consider concepts similar to these poles and, using, for example, the method of semantic differential, evaluate the proximity of all these concepts.
In addition, those respondents who oppose certain social groups to "cultured people" are of interest, among which homeless people and criminals stand out. It can be assumed that in the minds of these people, certain social groups have a certain measure of culture, therefore it would be of great interest to determine such a measure of culture in different social strata of the population using the same semantic methods. From this we conclude that within the same study both can and should be used various methods, approaches to obtaining and analyzing information.
The peculiarity of the data obtained using the method of unfinished sentences lies in the possibility of multiple access to primary information. At the same time, various bases for classifying textual information are used to solve various research problems. The MNE method has been successfully used to study social expectations, social identity. MNP is also used in mass surveys as a measurement method. Only in this case, one or more sentences are used.
Test of twenty "I" (TJT)
This method was developed in the 50s by M. Kuhn and T. McPartland to study the image of their own "I", for the study of self-definitions or self-identification of a person. The method of collecting information is quite simple. The respondent is offered a sheet of paper with the heading "Who am I", with twenty numbered lines. He is asked to answer, as to himself, "Who I" and write down your answers quickly enough; preferably in the form of nouns. In the order in which they come to mind. You don't have to worry about the logic or the importance of the answers.
The authors of this method found that responses are divided into four classes. Two of them relate to objective self-determinations (let's denote them K1 and K2), and the other two ¾ to subjective ones (let's denote them K3 and K4).
To class K1 include such self-determinations of the individual as a "physical" object (I am a man, I am a woman).
Class K2 constitute self-definitions representing the individual as social facility(I am ¾ member of society, I am ¾ student, I am ¾ music lover, I am ¾ scientist, I am ¾ teacher).
To the class of subjective self-determinations of short circuit include those that are associated, as it were, with socially relevant characteristics of behavior (I ¾ do not right person, I am ¾ a pessimist, I am ¾ lucky, I like to listen to music, I am a drinker in good company).
Class K4 form those self-definitions that are, to some extent, irrelevant to social behavior, and are also irrelevant to the self-identity task set by the test (to live ¾ to die).
Below is real data ¾ these are the answers of three students from a study we conducted among future linguists. Try to interpret these data, because these three situations will be encountered in research.
Answers from three students
It must be kept in mind that not all respondents give a complete answer. The number of completed lines in itself characterizes the personality of the respondent. As a rule, the respondent often lacks nouns for answers, and some of them have an “excess” of them. Respondents have or do not have self-definitions from the four indicated above. Based on this, we can proceed to a formal “description” of the respondent. Associate each with a set consisting of zeros and ones.
Theoretically possible 16 sets listed below:
0000 0001 0010 0011 0100 0101 0110 0111
1000 1001 1010 1011 1100 1101 1110 1111
If the respondent has self-definitions from all four classes, then he is assigned set 1111. If the respondent has only class self-definitions K2, then it is matched 0100. The number of possible sets is 2 4 =16.
In practice, not all sets are found. Such coding of textual information makes it possible to single out separate groups of respondents with the same self-identification structure. Thus, the sociologist finds typological groups, typological syndromes for the study of self-identification. Other ways of formalizing the textual information obtained through the test of twenty self-determinations are also possible.
What conclusions can be drawn from such a superficial examination of the methods LMS, MNP, test of twenty self-determinations.
1. All three ¾ methodological procedures in which it makes no sense to separate the collection technique, the measurement technique, the analysis technique. At the same time, in real research they can be used as a technique for measuring the properties of social objects, as a technique for analyzing social reality, and as a technique for collecting empirical data.
2. Each type of textual information requires a specific logical formalization. Only after that comes the turn for the formalization of the mathematical one.
3. All three can be used for different purposes and, in particular, for conducting a typological analysis of social phenomena.
1. Measurement ¾ component analysis of information associated with the procedure for obtaining initial data for analysis. Separate techniques for studying social reality are called both measurement techniques and analysis techniques (logical and analytical indices, ranking). Some measurement techniques (C. Osgood's semantic differential, the method of unfinished sentences) are also interpreted as approaches to the analysis of social reality. Measurement begins with a model for studying the properties of social phenomena.
2. Dimension ¾ is scaling (one-dimensional or multi-dimensional). Measurement ¾ is the procedure for obtaining a scale (Likert scale, Thurstone scale, Gutgman scale). Measurement ¾ obtaining the scale itself, i.e., a ruler with gradations (assuming the existence of a one-dimensional scale). Measurement ¾ is a diagnostic procedure.
3. If we correlate the methods of measurement with the types of empirical data, then we get the following conclusions. In the first type, we are talking about the metric level of measurement, and the measurement problem is reduced mainly to the formation of analytical indices and ranking. In the second type, measurement occurs as the coding of verbal judgments or the use of graphical scales. Finally, the problem of measurement fits in, is absorbed, conditioned by the different "origins" of texts.
B. P. Gromovik, A. D. Gasyuk,
L. A. Moroz, N. I. Chukhrai
The use of semantic differential in marketing research
Lviv State Medical University. Danila Galitsky
State University "Lviv Polytechnic"
IN modern conditions The need for marketing information is constantly growing, and marketing managers feel the lack of reliable, up-to-date and comprehensive data. To solve this problem, pharmaceutical enterprises should create a system for collecting the necessary marketing information - a marketing information system.
There are four main subsystems for collecting, processing, analyzing and researching marketing information, namely:
a subsystem of internal reporting of a pharmaceutical enterprise, which makes it possible to track indicators that reflect the level of sales, the amount of costs, the volume of inventory, cash flow, data on receivables and payables, etc.; | |
a subsystem for collecting current external marketing information, i.e. a set of sources and procedures used to obtain daily information about various market trends; | |
a marketing research subsystem for designing, collecting, processing and analyzing data that requires a special study of a specific marketing problem; | |
an analytical subsystem of marketing, consisting of a statistical bank and a bank of mathematical models and covering progressive tools for analyzing data and problem situations. |
If the external and internal information systematically accumulated in the marketing information system through market monitoring is insufficient, there is a need to conduct special studies of various marketing problems.
The process of marketing research takes place in several stages (Fig. 1).
Rice. 1. Marketing research process
At the first stage, it is necessary to determine the subject of research and goals, which should be clearly delineated and realistic.
Research objectives can be:
There are two types of marketing information collected in the research process:
Research mainly begins with the collection of secondary information. This stage is called "desk" research. Secondary information can be collected from internal and external sources.
In most cases of marketing research, after processing and analyzing secondary information, they move on to collecting primary data, which requires careful preparation. The plan for collecting information should determine, first of all, the method of research. The most used research methods are presented in fig. 2.
Rice. 2. Methods for collecting primary information
Observation is an analytical method by which the researcher studies the behavior of consumers, sales personnel; sometimes he acts as a participant in events (active observation).
The survey involves finding out the positions of people, their views on certain problems based on their answers to pre-prepared questions.
One type of survey is an in-depth interview, which is used to study the behavior of the consumer, his reaction to the design or advertising of the product.
In case of insufficient market research, it is necessary to:
Most often used:
- trade panel (especially the retail panel);
- consumer panel (of end consumers or consumer organizations).
Experiment - a method by which you can study (find out) the reaction of the studied group of people to certain factors or their change. The experiment aims to establish causal relationships between the variables under study by testing a working hypothesis.
Imitation - a method based on the use of computers and the study of relationships between various marketing variables on the appropriate mathematical models, and not in real conditions. It is used quite rarely.
The most common method is the survey, which is used by about 90% of marketing research.
As a rule, a common tool for collecting primary data is a questionnaire. When developing questionnaires, two types of questions are used: open and closed. An open question gives the respondent the opportunity to answer in their own words. The answers to them are more informative, but they are more difficult to process.
A closed question contains possible answers and the respondent chooses one of them. Closed questions may take different forms. The most common questions are alternative questions (assumes “yes” and “no” answers) and questions with selective answers. Quite often, researchers use different scales, in particular:
The stages of marketing research using the semantic differential are shown in fig. 3.
Rice. 3. Stages of marketing research using semantic differential
At the first stage, it is necessary to choose a comparison base, i.e. a competitor's product that contains the greatest threat to the enterprise under study and is the most representative on the market. Further, the consumer characteristics of this category of goods are determined, which are most important for the target group of consumers under study, and a system of evaluation of these characteristics is selected. After that, a questionnaire is developed to build a semantic differential. The next step is the questioning of consumers-respondents, i.e., the construction of semantic differential curves by them, guided by the perception of the characteristics of the product under study, the basic product-competitor and the hypothetical ideal product. The marketing research is completed by the construction of average curves based on the opinions of consumers and the analysis of each consumer characteristic of the studied goods.
As an example, the object of marketing research was the shampoo "Magiya trav" produced by the Nikolaev pharmaceutical factory and JV LLC "Magiya trav". Shampoo "Elseve" produced by the French company "L'oreal" was chosen as the base for comparison.
These goods were examined according to 10 consumer characteristics, which were evaluated on a 10-point scale (table). Respondents rated each position of the questionnaire with the corresponding score for the Magic Herb, Elseve shampoo and the ideal shampoo that they would like to purchase.
Table. Semantic differential of consumer characteristics of shampoos "Magiya herbs", "Elseve" and ideal shampoo
Based on the data obtained, average profiles of three curves were built, which reflect the average subjective perception of consumer characteristics of the studied goods and the vision of an ideal shampoo.
Analyzing the curves (table), it should be noted that the studied shampoo "Mag_ya herbs" satisfies the target consumers in terms of the following characteristics: pleasant smell; the effect of purity and silky shine; relatively well-known brand of goods and the presence of natural ingredients; price (lower than Elseve shampoo).
At the same time, consumers are not entirely satisfied with the packaging of Magic Herb shampoo, in particular, its design and convenience, as well as the absence of a conditioner in it. Therefore, we can recommend the manufacturer to pay more attention to improving the packaging and combining the shampoo with other components (conditioner, keratides, etc.). Attention should be paid to the availability of a sufficient amount of shampoo in the retail network as a factor in the availability of its purchase.
Thus, the use of the semantic differential in marketing research provides a thorough and visual differentiation of the characteristics of the compared goods. In addition, it helps to identify the needs of various categories of consumers before choosing a place for a product on the market, since the consumer perceives any product as a set of certain characteristics and, depending on their optimal set, gives preference to one product over another.
Literature
- Kovalenko M. // Business Inform.- 1997.- No. 1.- P. 59–62.
- Kutsachenko E. //Business.- 1999.- No. 31 (342).- P. 40–41.
- Mnushko Z. M., Dikhtyarova N. M. Management and marketing in pharmacy. Part II. Pharmaceutical Marketing: Pdr. for farm. universities and faculties / Ed. Z. M. Mnushko.- Kharkiv: Osnova, in-in UkrFA, 1999.- S. 237–241.
- Starostina A. O. Marketing research. Practical aspect. - K.; M.; St. Petersburg: View. dіm "Williams", 1998.- 262 p.
Instruction examples Report structure semantic differential. Description.
Abstract source: . Poll // Social psychology. Workshop: Proc. allowance for university students / Ed.
semantic differential can be defined as a method quantitative And quality value indexing. What does it mean?
According to C. Osgood, the semantic differential (SD) method allows you to measure the states that arise between the perception of a stimulus-stimulus and meaningful work with them. connotative points to something subjective, individual and value, is opposed to denotative - objective, interpersonal, cognitive.
We also recall that SD is one of the ordinal scale options. According to the classification of S. Stevens, the scales are divided into non-metric (nominal and ordinal) and metric (interval and ratios). Being a method of experimental semantics, SD along with other methods (for example, associative experiment, subjective scaling) is used to construct subjective semantic spaces, is widely used in sociology, general and social psychology. Appeal to him in psychological research justified when it comes to, for example, emotional attitude of the individual to certain objects, studying stereotypes, social representations, social categorization, attitudes, value orientations, subjective-personal meaning are considered, and implicit theories of personality are revealed. SD is classified as a method case studies, because it allows you to penetrate into the unique context of the life of the individual.
SD procedure
The method was developed by a group of American researchers led by C. Osgood, who considered it as a combination of controlled association and scaling procedures. For differentiation, it is proposed concept (a number of concepts), as well as a set of bipolar scales given by adjectives. The respondent must give an assessment of the differentiable object on each of the proposed bipolar seven-point scales. In response to the word, the respondent has a certain reaction, which reveals a certain similarity with the behavioral reaction, a kind of readiness for behavior, something mediating behavior. Respondent's associations with the stimulus guided by predetermined bipolar scales.
Functions these scales are as follows: firstly, they help verbalize a reaction to one stimulus or another; secondly, they contribute concentration on certain properties of this stimulus that are of interest for research; finally, with their help, it is possible to compare the estimates given by different respondents to different objects. Let's take a look at the use bipolar scales for estimating the object of interest to us. This is an easy and economical way to get information about human reactions.
The idea of using bipolar scales goes back to the early studies of synesthesia, conducted by Osgood, together with T. Karvoski and G. Odbert. Osgood proposes to understand synesthesia as a phenomenon that characterizes the experience of individuals, in which certain sensations belonging to one feeling or modality are combined with certain sensations of another modality and occur every time when there is an incentive corresponding to another modality (it is worth remembering, for example, the synesthesias of A. Scriabin, V. Kandinsky, V. Nabokov).
In his research on synesthesia, Ch. Osgood looked for connections between synesthesia, on the one hand, and thinking and language, on the other. results experimental work, supported by the analysis of facts from cultural anthropology, led to the conclusion that the images found in synesthesia are closely related to language metaphors, and all this is semantic relations. Metaphor in language, as well as musical-color synesthesia, can be described "as a parallel alignment of two or more dimensions of experience", which are defined using pairs of opposite adjectives. It is the appeal to the mechanisms of synesthesia that makes it possible to explain metaphorical transfers in such statements as “ sour face", « bad character".
With the help of some bipolar scales, profiles of social stereotypes. Respondents in several samples were asked to evaluate such objects as a pacifist, a Russian, a dictator, and neutrality on bipolar scales. During the Second World War, researchers recorded a change in the structure of social stereotypes (or, as C. Osgood writes, a change in the meaning of social signs) since the US entered the war.
It also turned out that when evaluating objects, bipolar scales (decent - dishonest, high - low, good - evil, useful - useless, Christian - anti-Christian, honest - dishonest) found a high correlation - 0.9 and higher, becoming an evaluation factor.
Scales (strong - weak, realistic - unrealistic, happy - unhappy) did not show correlations with rating scales, which allowed the researchers to talk about existence and other dimensions of the semantic space.
Semantic differentiation, according to Osgood, involves a consistent location of a concept in a multidimensional semantic space by choosing one or another value between the poles on the scales.
The difference in the meanings of two concepts is a function of the multidimensional distance between the two points corresponding to these concepts.
The scales proposed for evaluating objects, and instructions can look like this (see the example and instructions in paragraph 3 of this. document) .
The use of such a scale makes it possible to directly measure the reaction of an individual, i.e., to identify a qualitative parameter (in this case, choose between "good" or "bad"), as well as to determine the intensity of this reaction (from low to high severity).
The scales are presented in random order, i.e. scales of one factor should not be grouped into blocks. The poles of the scales should not create in the respondent the attitude that the left pole always corresponds to a negative quality, and the right pole - to a positive one.
Space compression and factors :
Evaluation factor combined the scales bad - good, beautiful - ugly, sweet - sour, clean - dirty, tasty - tasteless, useful - useless, good - evil, pleasant - unpleasant, sweet - bitter, cheerful - sad, divine - secular, pleasant - unpleasant, fragrant - smelly, honest - dishonest, fair - unfair.
strength factor : big - small, strong - weak, heavy - light, thick - thin.
activity factor : fast - slow, active - passive, hot - cold, sharp - blunt, round - angular. The evaluation factor in this study played the main role, it explained 68.6% of the total variance, while the remaining factors - 15.5 and 12.7%.
These three independent factors were obtained in Numerous studies conducted in various cultures, among subjects with different levels of education, on the material of various objects (concepts, as well as stories and poems, social roles and stereotypes, images, colors, sounds, etc.)
However, the procedure factor analysis is not the only way analysis of data obtained using method C also offers a formula by which distance between scaling objects, i.e., two points in the semantic space. After all, scaled objects can be represented in the form semantic profiles
When scaling narrow set of concepts going on transformation of three-dimensional space"estimation - strength - activity", i.e. independent orthogonal factors cease to be such.
For example : C. Osgood invited respondents to evaluate 20 concepts: 10 politicians (including R. Taft, W. Churchill, I. Stalin, G. Truman, D. Eisenhower) and 10 other realities (US policy in China, socialism, state price control, use atomic bomb, UN, etc.) on 10 bipolar scales (including: wise - stupid, clean - dirty, dangerous - safe, unfair - fair, strong - weak, idealistic - realistic, etc.). As a result, instead of the three-dimensional space ≪assessment - strength - activity≫, a one-dimensional continuum was obtained with poles ≪ benevolent dynamism ≫ and ≪spiteful impotence≫.
SD as a way to measure attitudes .
Let's look at studies where the SD method has been used to study attitudes. Let us pay special attention to how the data obtained with the help of SD is analyzed. In the work of C. Osgood, devoted to the study of attitudes towards representatives of different races, respondents (white and black students) were asked to evaluate a number of concepts (including concepts indicating belonging to a race) on 12 bipolar scales (6 scales according to the factor “assessment ”, 3 scales for the “strength” factor, 3 for the “activity” factor). After calculating the average values for each concept on the scales of three factors, calculating the semantic distances between the scaled concepts for different groups of subjects, it turned out that white respondents have positive attitudes towards those who belong to the Caucasoid race, less positive attitudes towards representatives of other races.
An analogy was also observed in the assessments of colors by this group of respondents. Interesting changes in ratings the concept of "man » depending on the adjective, denoting color. For white respondents, the adjective dominates the noun, and " connotative meaning the concept of a black person is rather black human, not black Human". Black respondents gave similar assessments of colors. Received the most positive rating White color, then yellow, red, finally brown and black. However, concepts indicating race were evaluated differently by this group. The most positive assessment was received by the concept denoting a representative of the Negroid race, the least positive - by the representative of the Caucasian race. For white students, the concept of "representative of the Caucasian race" found more similarity with the concept of "citizen", and not with the concept of "foreigner", "friend" rather than "enemy", in the case of the concept of "representative of a race" the similarity was the opposite: it was more similar to the concept of "foreigner" and not "citizen", with the concept of "enemy" and not "friend". For this group of respondents, the concept of "person" is most similar to the concept of "representative of the Caucasian race" and least similar to the concept of "representative of the Negroid race". In the group of black respondents, the opposite results were obtained.
Now let's stop at limitations of this method . Its main limitation is that we are dealing with declared verbal reaction of the respondent. The placement of stimuli in the semantic space turns out to be distorted under the influence of, for example, social desirability or other mechanisms of this kind. To overcome this shortcoming, one can use modifications of the SD method, for example, non-verbal SD, in which the effect of conscious correction of evaluative reactions is reduced.
Stages of work on compiling a private SD (Step 1-2 have already been completed by you / in theory, should have been completed by the last lesson)Let's imagine that you are conducting a study, as a result of which you want to compare the perceptions of a successful woman in men and women. The logic of your research should be built as follows:
STAGE 1: It is necessary to conduct an initial survey of the subjects. A sample instruction might read as follows: “Name 10 (15 or whatever number you need) characteristics that describe a successful woman.
If you are comparing the representations of men and women, then you need to interview at the first stage the SAME number of both men and women.
STAGE 2:
A. for each group of subjects: men and women SEPARATELY You counting the number of all mentioned characteristics. For example, “kind - 7 (occurs 7 times), beautiful - 9 (occurs 9 times), etc. After you have calculated the characteristics for men and women, you need to compare the data between groups.
B. As a result, you should get feature list, which are most often found in two groups of subjects: men and women. FREQUENCY characteristics are those that are found in more than 50% of the questionnaires (i.e. if at least half of your subjects mentioned any characteristic, it should be considered frequency). The bar for assessing the "frequency" of a feature can vary from 80% to 30%, depending on the characteristics obtained. If none of your characteristics scored more than 30%, then further construction of a private SD differential is meaningless. You must either add respondents or analyze the concepts you ask to describe.
IN. To the resulting list of characteristics, it is necessary pick up antonyms. For example, you received characteristics: kind, smart, rich, etc. Each characteristic should receive an antonym: GOOD - EVIL, SMART - STUPID, RICH - POOR. Antonyms are best selected using the dictionary of antonyms !!!
The resulting pairs of qualities will be scales for evaluation. Scales can be 5-point or 7-point (rarely 9-point or 11-point): For example:
Good 3 2 1 0 1 2 3 Evil
Smart 3 2 1 0 1 2 3 Stupid
Rich 3 2 1 0 1 2 3 Poor, etc.
Do not forget : shuffle the poles of the scales (so as not to create settings)
determine for yourself the poles of each scale, i.e., assign "internal" numerical values that the respondent will not know, only you (for later interpretation)
STAGE 3.
A. Drawing up instructions for the SD and choosing the concepts that respondents will evaluate (for example, “successful woman”, “unsuccessful woman”, “woman”, women of various professions, etc.). The object can be one or more, depending on the purpose and topic of the study.
Additional note: if SD is a separate study, then this questionnaire must also comply with the requirements applicable to the questionnaire. That is, contain: a greeting, a legend, instructions, the CD itself, a passport, Gratitude.
B. Questioning the subjects using the received questionnaire. The subjects must be DIFFERENT than in the first stage.
STAGE 4: Processing received data.
A. Entering the received estimates in the table. Processing using factor analysis.
B. Determine the required number of factors. Select factors according to the obtained factor loadings.
IN. Give the factors meaningful names and psychological interpretation.
Instruction optionssee separate file ("SD_options_instructions")
Report Structuresee separate file ("Poll_report_structure")
A person, perceiving any object, does it through two channels. First, it gives the object a denotative meaning, i.e. the meaning that he learned in the process of education. For members of the same community, the denotative meaning of one object is the same. For example, apples are useful for humans, contain many vitamins and have a good effect on complexion. Such a meaning will be given to the apple by those communities in which great importance given to a healthy lifestyle. Another community may have a different perception of the apple: an apple is a fruit that needs to be stored in the basement in boxes with straw and preferably consumed before spring, because. in the spring they will start to deteriorate. In both the first and second examples, a person perceives the meaning of the object not through personal “communication” with the apple, but from the process of socialization.
In addition to the denotative meaning, each object for a person has a connotative meaning. This meaning is personal, obtained through direct experience. If one fine sunny day a heavy apple fell on my head, I lost consciousness, and when I woke up I realized that I was lying in a pile of cow dung, then for the rest of my life I would avoid clusters of large apples on trees. In the above example, the experience of "communication" with an apple is very vivid. Usually the connotative meaning is more hidden.
I will give other examples of connotative meanings. The rector of the university can be assessed by his students as a solid and cold-tempered man. This does not mean that the density and temperature of the rector's body differs from the average values of other people outside the normal range.
In other words, the connotative meaning is an emotion in relation to the object being evaluated.
And what about the semantics? We introduce the definition according to Tolstova. Semantics is a branch of linguistics and logic that studies the problems of meaning, meaning and interpretation of signs and sign expressions. Accordingly, psychosemantics is the study of a person's psychological perception of the meanings and meanings of various kinds of objects. Psychosemantics includes such methods as semantic differential, repertory grids, etc.
The task of psychosematics is very interesting - the construction of the semantic space J. I.e. system of latent factors within which a person works. Why did you bypass the puddle in front of the entrance on the right side this morning, although it was more convenient on the left?
Why does sociology need SD? For example, a sociologist may be concerned with identifying types of people with similar perceptions of objects. If the object is the advertised product, then for each individual type it is more efficient to make a separate advertisement with the desired perception J
A big plus of SD is that, using "hard" methods, it provides information about subtle psychological structures human perception of objects.
Semantic Differential Technique
What did Osgood suggest? The emotion of the meaning of the concept will be revealed if a person points to the position of the concept in question in the system of connotative features. Those. will indicate the location of the object in the "emotional" coordinate system. For example, evaluate a political leader: is he warm or cold, fluffy or prickly?
So we are preparing several pairs of emotions (connotative features). Couples, of course, contain opposite emotional colors: sour-sweet, black-white, good-evil. Each pair contains several gradations. If you want to use factor analysis in your analysis, you need data defined by an interval scale. For this, there must be seven gradations (the more gradations, the more your scale shifts from ordinal type to interval type).
Rate Vasya Pupkin | ||||||||
light | -3 | -2 | -1 | 0 | 1 | 2 | 3 | dark |
cold | -3 | -2 | -1 | 0 | 1 | 2 | 3 | warm |
calm | -3 | -2 | -1 | 0 | 1 | 2 | 3 | anxious |
fog | -3 | -2 | -1 | 0 | 1 | 2 | 3 | clear |
useful | -3 | -2 | -1 | 0 | 1 | 2 | 3 | harmful |
sad | -3 | -2 | -1 | 0 | 1 | 2 | 3 | glad |
solid | -3 | -2 | -1 | 0 | 1 | 2 | 3 | unsteady |
false | -3 | -2 | -1 | 0 | 1 | 2 | 3 | true |
peaceful | -3 | -2 | -1 | 0 | 1 | 2 | 3 | warlike |
nonsensical | -3 | -2 | -1 | 0 | 1 | 2 | 3 | reasonable |
Rate Vova Golikov | ||||||||
light | -3 | -2 | -1 | 0 | 1 | 2 | 3 | dark |
cold | -3 | -2 | -1 | 0 | 1 | 2 | 3 | warm |
calm | -3 | -2 | -1 | 0 | 1 | 2 | 3 | anxious |
fog | -3 | -2 | -1 | 0 | 1 | 2 | 3 | clear |
useful | -3 | -2 | -1 | 0 | 1 | 2 | 3 | harmful |
sad | -3 | -2 | -1 | 0 | 1 | 2 | 3 | glad |
solid | -3 | -2 | -1 | 0 | 1 | 2 | 3 | unsteady |
false | -3 | -2 | -1 | 0 | 1 | 2 | 3 | true |
peaceful | -3 | -2 | -1 | 0 | 1 | 2 | 3 | warlike |
nonsensical | -3 | -2 | -1 | 0 | 1 | 2 | 3 | reasonable |
Vasya Pupkin's score | ||||||||||
light dark | cold - warm | calm - anxious | hazy - clear | useful - harmful | sad - happy | hard - shaky | false - true | peaceful - militant | senseless - reasonable | |
rep1 | -2 | 2 | 2 | 2 | 0 | -3 | 0 | -3 | 0 | 0 |
rep2 | -3 | -1 | 1 | 1 | -1 | -3 | -3 | -1 | -1 | -1 |
rep3 | 1 | -3 | -1 | -2 | 0 | -1 | 1 | 2 | -3 | 2 |
rep4 | -1 | -2 | -2 | -2 | -3 | -1 | -2 | -2 | -1 | -3 |
rep5 | -1 | -2 | -2 | -3 | -3 | -1 | -2 | 0 | -1 | 1 |
Vova Golikov's score | ||||||||||
rep1 | -2 | -2 | -1 | 0 | 0 | -2 | -2 | -2 | -1 | -3 |
rep2 | -1 | 0 | 1 | -3 | -1 | -1 | 2 | -1 | 0 | -2 |
rep3 | -2 | 2 | 1 | 2 | 0 | 1 | 2 | -3 | 1 | 2 |
rep4 | 0 | 0 | 2 | -3 | -3 | 0 | -1 | -2 | 0 | -3 |
rep5 | -2 | 0 | -3 | -1 | -2 | -1 | 1 | 1 | 0 | -2 |
Osgood found that in most cases, one of the three is hidden behind any of the connotative pairs. options: strength, assessment (attitude), activity. In other words, if we take an object, let the respondents evaluate it by a hundred similar pairs, and then carry out a cluster analysis of all these pairs, then we will see that all pairs are divided into three groups: strength, assessment, activity. Those. when we perceive any object of reality, we "attribute points" to this object according to three characteristics: strength (strong-weak), evaluation (bad-good) and activity (fast-slow).
Each researcher can create his own scale, but it is hardly worth it. It is better to choose a scale from among standard scales that are original in the sense that they have their own own name, are widely used, are included in the most common system of scales. They are called original. Further, four discrete rating scales are considered, ratings - Likert, semantic differential, graphic rating and Stepel, as well as a scale with a constant sum and a rank scale.
Likert scale based on degree choice agreement - disagreement with some specific statement. In fact, one pole of this essentially bipolar ordinal scale is formulated, which is much easier than naming both poles. The formulation of the statement may correspond to the ideal level of some parameter of the object. When describing the higher educational institution we can consider the following properties: qualified teaching staff, equipment auditor fund technical means, modernity and regularity of updating training courses, availability e-leming V educational technologies, the level of culture, image and reputation, the contingent of students and many others. The wording of statements can be as follows: the teaching staff of this university is very qualified; the university has a very high level of application of modern teaching aids; students who strive for knowledge study at this university; Graduates of this university are highly valued in the labor market.
When applying the Likert scale, five gradations are usually considered. An example of using the Likert scale in the questionnaire is shown in fig. 8.1. In other words, the questions are formulated in the format of a Likert scale. The respondent is asked to tick one of the five boxes.
Rice. 8.1.
At the same time, the quantitative assessment itself is not required from the respondent in this case, although more often points can be immediately affixed next to the names of the gradations. As can be seen from fig. 8.1, the degree of agreement-disagreement with each statement may have the following gradations: strongly disagree (1 point), disagree (2 points), neutral (3 points), agree (4 points), definitely agree (5 points). Here, in parentheses, the most commonly used version of the scale digitization is given. It is also possible that a higher score (5 points) corresponds to the "strongly disagree" gradation.
Semantic differential and graphic rating scale
Semantic differential scale implies the presence of two polar semantic meanings (antonyms) or antonymic positions, between which is located odd number gradations. In this sense, the scale is bipolar. As a rule, seven gradations are considered. The middle position (medium gradation) is considered neutral. Digitization of scale gradations can be unipolar, for example in the form "1, 2, 3, 4, 5, 6, 7", or bipolar, for example in the form "-3, -2, -1, 0, 1, 2, 3".
Usually the poles of the scales are given verbally (verbally). Examples of scales with two poles are as follows: "soothing - invigorating" or "compact - voluminous". Along with verbal semantic differentials, non-verbal semantic differentials have been developed that use graphic images as poles.
Examples of verbal semantic differentials are given in fig. 8.2.
Rice. 8.2.
The semantic differential resembles the Likert scale, but has the following differences: 1) both polar statements are formulated instead of one; 2) instead of the names of intermediate gradations, a sequential graphic arrangement of an odd number of gradations located between the extreme values of "good - bad" is given.
Semantic differential method (from the Greek. sematicos - denoting and lat. differentia- difference) was proposed by the American psychologist C. Osgood in 1952 and is used in studies related to human perception and behavior, with the analysis of social attitudes and personal meanings, in psychology and sociology, in the theory of mass communications and advertising, and in marketing.
It can be considered as an analogue of the semantic differential scale. The rating scale is implemented in such a way that each property is assigned a line, the ends of which correspond to polar statements, for example: "not important" and "very important", "good" and "bad" (Fig. 8.3).
Rice. 8.3.
The fundamental difference between the compared scales is that the semantic differential is a discrete scale, and, as a rule, it has seven gradations, and the graphic rating scale is continuous.
- So, when characterizing the exterior of certain brands of cars, it is sometimes said that brutality is inherent in it. There are more simple examples– ergonomics and controllability, when it is difficult to meaningfully name the second pole.