The semantic differential (SD) is one of the projective methods of sociology, based on the achievements of psychosemantics, and was developed by a group of American psychologists led by Charles Osgood in 1952. It is used in studies related to human perception and behavior, with analysis social attitudes and personal meanings. The SD method is a combination of the controlled association method and scaling procedures.
Psychosemantic methods transfer information from the cognitive level (and the research task is always formulated in its terms) to the affective level, where this information is encoded not by linguistic forms, but by various sensations.
The method of semantic differential is based on the phenomenon of synesthesia (thinking by analogy, when some sensory perceptions arise under the influence of others) and is an operational way of “capturing” the emotional side of the meaning perceived by an individual in objects. SD makes it possible to identify unconscious associative connections between objects in people’s minds.
The SD method makes it possible to find a system of latent factors within the framework of which an individual evaluates objects. Essentially, semantic space is a research model of the structure of individual consciousness, and the task is to determine where the object being studied is located in this space.
The tested objects (name, brand, packaging, etc.) are assessed on a number of bimodal seven-point scales, the poles of which are usually specified verbally using antonyms: good - bad, warm - cold, active - passive, etc. It is assumed that a person is able to evaluate the object being studied by correlating the intensity of internal experience about the object with a given rating scale. The divisions of the scale record different degrees of a given quality of an object. Scales that correlate with each other are grouped into independent factors that form a semantic space.
Along with verbal ones, non-verbal semantic differentials have also been developed, where graphic oppositions, paintings, and photographic portraits are used as poles of scales.
Research often uses monopolar scales, with the help of which objects are assessed according to the severity of one property: how good the object is, how warm it is, etc. In the case of bimodal scales, the respondent evaluates where object “A” is located for him on the “expensive - cheap” scale, and with unimodal scales, he evaluates how “expensive” the property is inherent in object “A”. The use of unimodal scales is due to the fact that often antonymous adjectives are not actually complete opposites - bad is not always bad.
In the classic version of Charles Osgood, exclusively connotative features were used as scales, which reflected not the objective properties of the assessed object or concept, but the subjectively significant aspects of the object or concept for the respondent.
In marketing research, a recognized tool for studying the image of a corporation, brand or product is denotative scales, which do not always consist only of antonym adjectives, but are, as a rule, phrases, phrases that express expectations, product characteristics, both negative and positive. Similar products from different manufacturing companies can be evaluated on a “worth the money” scale, for example, banks - according to the level of reliability, profitability, etc.
To preserve the “spirit” of the method and capture the still affective elements of the attitude, a set of scales (15-25 scales) is used. The result of the technique is not directly calculated average values of objects on each of the scales, but latent factors obtained during a special analysis procedure, on the basis of which the semantic space of perception of objects is formed and a map of their relative positions is constructed. It is important to select a sufficient number of scales and test them on experts or conduct an association experiment on representatives of the target group in order to avoid the danger of researcher subjectivity when choosing scales.
SD scales do not describe reality, but are a metaphorical expression of the subject’s states and relationships (the instructions that respondents receive call for: “When making assessments, be guided by your own feelings, not knowledge”). In the resulting space of affective meanings, a convergence of concepts to which a person reacts in a similar way and a separation of concepts that have different emotional backgrounds are recorded. The distance between concepts is expressed by a certain number, which in general view allows you to distinguish between assessments of: a) the same concept by different individuals (or different groups); b) different concepts by the same individual (or group); c) one concept by the same individual (or group) at different times.
The number of identified factors corresponds to the structure of the emotional perception of a given class of objects, so, for example, when evaluating a bank, only two factors can be identified: reliability and profitability, while a car can be assessed according to the criteria of “fashionability, style,” “prestige, status,” “ prices”, “operational efficiency”, “after-sales service networks”, etc.
The procedure for developing a semantic differential methodology within a specific research project usually consists of the following steps:
Formation and testing of a list of adjectives, statements to describe the objects being tested (names, concepts, types of packaging, brands, etc.). The level of awareness at which the respondent will evaluate the measured object depends on the selected features. By focusing on denotative scales, we expand the semantic space, increasing information about objects and inevitably losing information about subjects, which is not so critical in marketing research.
Mathematical processing of the resulting data matrix: object - respondent - scale. The procedure commonly used factor analysis, which makes it possible to identify latent assessment criteria into which the initial scales are added. It is important to note that to obtain significant results, relatively small samples - 30-50 people - are sufficient, due to the fact that the unit of analysis is not the respondent, but the ratings he gives to objects. Considering that each of the 30-50 respondents evaluates 7-10 objects on 15-25 scales, the total sample size turns out to be quite sufficient to draw statistically significant conclusions.
Placement of evaluated objects in the constructed semantic space, analysis of the resulting distribution. Estimation of the distance between the tested objects and the ideal object (for example, ideal yogurt, car, “myself,” etc.), to determine the “positive” poles of the factors. For example, if we received the factor “fashionability, stylishness, brightness” of a car, then it is important to understand whether the high ratings of our brand on this factor are positive for the target audience or not. Perhaps the ideal car for them is a reliable, conservative “iron horse”, economical in fuel consumption and without any special quirks in design.
Stage 1 Formation and testing of a list of statements.
The tools used in the semantic differential technique usually consist of a table the following type: scales are placed along the rows, and evaluated objects are placed in the columns. The instructions offered to the respondent are formulated approximately as follows: “Please rate the characteristics of each of the brands “…” on a scale from 0 to 5, where 0 means no such property, and 5 means the property is expressed to the maximum. In the “ideal ...” column, write down what properties good ... should have, using a scale from 0 to 5, where 0 - such a property should not exist, and 5 - the property should be inherent in the product to the maximum extent.”
Considering that a completely sufficient sample for a homogeneous group of respondents within the framework of this technique is 30-50 people, it is often convenient to collect information in parallel with a focus group study. Considering that usually the launch of a new brand, name, or packaging is accompanied by a series of focus groups, then during 3-5 groups it is possible to collect 30-50 questionnaires. This sample size turns out to be quite sufficient to supplement the conscious, rational information provided by respondents with assessments of the affective elements of the attitude, i.e. collect extraconscious, emotional, irrational data that the semantic differential technique allows you to obtain.
Figure 4 - Example of a semantic differential table for object evaluation
Stage 2. Mathematical processing of results and their interpretation
The SD technique makes it possible to quite clearly process the results and interpret them using the simplest statistical characteristics. The average value of the measured value, standard deviation, and correlation coefficient are proposed as such characteristics. Primary processing of the results consists of compiling statistical series measured value for each object under study. Then the average statistical value of the measured value for the sample and a measure of the unanimity of estimates, expressed by the standard deviation, are calculated. Once the average ratings of each object for the three measured indicators have been identified, it is interesting to trace their interdependence. Thus, the algorithm for mathematical processing of SD results is as follows:
Step 1. Drawing up a statistical series in the form of a table.
X i - assessment of a certain quality of an object on a seven-point scale;
n i - frequency of the value X i, i.e. how many times was the score X i given when assessing the object according to the parameter under study by all respondents in the aggregate.
Step 2. Calculation of the average value.
If K respondents participated in the survey, then the average value is calculated using the formula:
n=M*K, since the quality under study is assessed K by respondents in the developed form M times (in M pairs of antonymous adjectives). The average value of X serves as an indicator of the overall assessment of a given quality of an object by the entire class, being at the same time a fairly objective characteristic, since it allows one to neutralize the influence of subjective factors (for example, the bias of individual respondents in relation to a given object at the time of the survey).
Step 3. Calculate the standard deviation.
The standard deviation serves as an indicator of the measure of dispersion of the values of a quantity around its average value X, i.e. measures of unanimity and cohesion of respondents in assessing a given quality of an object. The standard deviation is calculated as the square root of the variance y x = vD x, where the variance D x, in turn, is calculated by the formula:
The described three steps of mathematical processing of diagnostic data reveal the picture of the respondents’ perception of the objects under study. This allows you to clearly present the results of the analysis.
The data obtained after the above-described processing can be compared with each other by calculating their correlation. This stage of processing aims to establish to what extent the respondents’ attitude towards the object is related to its individual characteristics.
Step 4. Calculation of the correlation of the obtained estimates.
When determining the correlation coefficient, firstly, the average value of the ratings of each of the indicators for all assessed objects is calculated. Let's say a respondent evaluates n objects. Based on activity, the 1st object was assessed by the average value of A j. Then the average score of indicator A of all objects:
Average P indicator score:
Then the correlation coefficient of A and P r A,P:
(covariance); , - standard deviations of the values A j and O j from their average values, which are found as follows:
As a result of calculating the correlation of ratings, one can clearly see the psychological mechanism for constructing the relationship of respondents’ ratings to the objects under study.
Stage 3. Presentation of the location of the tested brands in the semantic space.
After the mathematical processing stage, several main factors can be identified and the location of the tested brands in the semantic space formed by the identified latent factors can be presented.
The results ultimately turn out to be quite clear and quite easy to interpret: the figure shows that an ideal product should have high quality and an affordable price (for the sake of clarity, the example has selected quite obvious properties). In terms of quality factor, brands 1 and 2 are closest to the ideal product, and in terms of price factor, brands 4 and 5. By assessing the set of criteria, we can conclude that brand 1 is closest to the ideal.
In a similar way, you can test, for example, variants of names, choosing names that evoke the most positive emotions, at the same time, associated with a specific product and evoking an image, association with the corresponding valuable qualities.
Interesting results can be obtained by comparing products that are not competing with each other, but have a similar basis, which makes possible comparison and helps to identify new positively assessed qualities of a product or brand and transfer them to a new product field (invention for application).
For example, an assessment of plastic cards in general, in order to understand what properties of fuel plastic cards need to be developed, and the use of which would help in capturing the fuel card market.
The semantic differential technique allows, when studying a brand, to identify an emotional attitude towards it (the affective component of the attitude), not burdened by rationalizing motives (the cognitive aspect). Identify how the potential consumer feels about the brand, i.e. predict his real behavior, and not words about actions.
The semantic differential allows you to draw statistically significant conclusions on small samples (sufficient material can be collected from 3-5 homogeneous focus groups) due to the fact that the unit of analysis is not the respondent, but the assessment (on average, each respondent evaluates 7-10 objects of 15 -25 scales, i.e. gives 100-250 ratings).
The SD method allows us to identify the structure of latent factors, criteria on the basis of which respondents construct evaluations of various brands. Accordingly, using the SD method, it is possible to construct a map of the placement of brands of interest in the structure of factors, thereby obtaining a visual, relatively easily interpreted research result.
The use of an “ideal” object in the SD methodology, along with the brands being tested, allows us to determine the desired directions of development, possible threats to the brand, and the most significant (although sometimes unconscious by the consumer) properties of the product.
The use of SD methodology in marketing research allows one to evaluate a brand and its elements (name, packaging, corporate identity, etc.), obtaining statistically significant estimates of the deep structures of consumers’ consciousness in the course of a relatively inexpensive and small-scale study.
One of the most widely used techniques of this type is the so-called “semantic differential” (C. Osgood, 1952). This is essentially not one, but a whole family of methods, a whole technology. In working with children, it can be used to study the level of formation and integrity of the child’s ideas about the world (cognitive diagnostics), and as a projective technique - for the study of personal attitudes and emotional relationships child to a certain range of objects.
The usual “semantic differential” (SD) is several seven-point scales applied horizontally on one form (answer sheet). The seven gradations are usually designated in words, just like the poles of the scale. Here is an example of a form:
Object "SUN"
ACTIVE strongly moderately weakly not at all weakly moderately strongly PASSIVE
EVIL strongly moderately weakly not at all weakly moderately strongly KIND
SOFT strongly medium weak not at all weak medium strong HARD
LIGHT strongly moderately weakly not at all weakly moderately strongly HEAVY
COLD very moderately weak not at all weak moderately strongly WARM
The subject's task is to record his assessment in the form of an assignment to a certain pole of the scale with a certain gradation. The selected gradation must either be underlined on the form or circled. Thus, each line of the form must contain a mark indicating the test subject’s answer.
As we see, in comparison with the Dembo-Rubinstein technique, the “semantic differential” is more perfect in that it is protected from the so-called “positional tactics”. Here, positively colored characteristics are placed not at the same pole of each scale (at the top), but at different ones - sometimes on the left, sometimes on the right.
As a result of filling out the SD form on the response sheet, a subjective semantic profile of the scaled object appears. It is easier to see if you connect all the marks with a single broken line.
When processing SD results, two approaches are possible: either analyze only profiles, or build a so-called “semantic space”.
Let us explain how you can act in the first case. Let’s say we are conducting SD for the purpose of career guidance consultation and asking a high school student to list the names of various professions that, as it turns out from a conversation with him, are potentially attractive to him. But which is the most attractive of them? To answer this question, the student is asked to scale, in addition to the names of specific professions, also a special ideal object - “the best profession for me.” After this, a comparison is made of all the profiles of real professions and the profile ideal profession(we will omit the formula for calculating the similarity measure here; the main thing here is to understand the general meaning of the method). And that real profession, the profile of which reveals more similarities with the “ideal” one, is declared as a result the best subjective choice.
[Note. It is clear that for different subjects this choice may be different, not only due to divergent ideas about real professions, but also due to differences in the profile of the ideal profession: some strive more for the “strong” (or, as in modern youth slang, “cool”) activities, others - to complex and interesting, others - to calm and kind, etc.].
“To build a “semantic space”, scale ratings are combined on related scales included in the same coordinate (factor) of the semantic space. As shown by numerous factor-analytic studies of foreign and domestic psychologists (K. A. Artemyeva, 1980, E. F. Petrenko, 1979, 1988, A. M. Etkind, 1U79, L. G. Shmelev, 1983 and others), most rating scales are combined into three summary rating scales: “good - bad”, “strong - weak”, “active - passive”. For example, ratings on the "soft - hard" scale turn out to be psychologically equivalent to ratings on a scale of 4 "good - bad", and ratings on the " heavy light“are actually close to the estimates on the “strong-weak” scale. As a result of such recalculation (very similar to calculating the total score on a test, but only in this case not for the subject, but for the object), each object receives a value for three main semantic factors and can be displayed geometrically as a point in the three-dimensional space Score-Strength -Activity". The similarity of profiles is the proximity of certain points in the semantic space; it can literally be seen (visualized).
Using an example, the professions of “pilot” and “traffic police inspector” that are closest to the ideal of the considered professions for the subject are called “pilot” and “traffic police inspector”. After that, it remains to find out whether the subject has real professionally important qualities in order to qualify for mastery of these professions.
We especially note that SD is also used to diagnose self-esteem. It is enough to ask the child to evaluate himself on the same set of scales. As a result, the point “I” appears in the semantic space. The degree of removal of this point from the “ideal” - a measure of self-dissatisfaction.
We emphasize once again that for the successful use of SD, the material that the child scales must be age-appropriate to his range of interests, otherwise the child will simply put a more or less random pattern of marks on the form.
WITH younger schoolchildren and for preschoolers, SD is best carried out individually and orally, that is, all marks on the form must be entered by the experimenter himself - based on the child’s oral responses. It is better to use fairy-tale or cartoon characters as material (objects for evaluation). “Special research by V.F. Petrenko has shown the high effectiveness of this material in working with children. A special modification of SD allows fairy-tale characters to denote poles of scales rather than objects. In this case, parents, friends and teachers will be compared by the child with certain fairy-tale characters.”
Specific problems and difficulties of family identification or school adaptation can manifest themselves in SD as in any projective technique. For example, unloved elder sister may turn out to be close in semantic space to “Baba Yaga”, and the head teacher, whom the child is simply afraid of, to some kind of “killer robot”.
The semantic differential method was developed in the 50s by American scientists under the leadership of Charles Osgood. On this moment Many applications have been found for it in various fields. This method is a tool for studying the semantic spaces of the subject and serves to index meanings using bipolar scales defined by two opposite adjectives, between which there are three, five or seven gradations of the degree of occurrence of a given quality. Any object, phenomenon or feeling perceived by an individual evokes some kind of reaction in him, which can be successfully characterized using semantic differential methods. These methods allow you to see the image that appears in the recipient’s mind when evaluating an object.
Can be considered as a type of projective tests that allow us to take into account the fact that a certain stimulating situation acquires meaning not only due to its objective content, but also for reasons related to the characteristics of the recipient himself - inclinations, drives, beliefs - which he attaches to this situation . Simply put, the individual traits of the test subject seem to be projected onto situations, affecting the test results. This method allows you to measure connotative meaning - a state that follows the perception of a stimulus symbol and precedes operations with symbols. The connotative meaning is directly related to personal qualities recipient, such as social attitudes, stereotypes, etc. and is close to us in the concept of personal meaning.
As mentioned earlier, objects in the semantic differential method are assessed on a number of opposing (bipolar) graduated scales. The extreme values of these scales are antonyms. Evaluations of concepts on different scales interact with each other, which makes it possible to identify bundles of such strongly interacting scales and group them into factors. This mechanism, which explains the grouping of scales together, was considered by Osgood to be synesthesia. Synesthesia is a phenomenon when, when one sense organ is stimulated due to its specific sensations, sensations corresponding to another sense organ also arise. An example is the case when, looking at some object, some taste sensations may arise.
When the transition from features to factors is made, this is already the construction of a semantic space, which in some way is a metalanguage for describing meanings. Osgood, in his research, built a semantic space based on the gradation of various conceptual classes (for example, father, ice, table).
Three main factors have been identified
- "grade" ( light-dark, ppleasant-unpleasant);
- "force" ( durable-fragile, strong-weak);
- "activity" ( fast-slow,active-passive).
All these factors together form a semantic space.
The method of semantic psychological differential, developed by Osgood, makes it possible to study not only the meaning of words, but also their emotional connotation, since the identified factors made it possible to study in more detail the structure of a person’s (or group of people’s) thinking.
Numerous further studies in this area only confirmed the universality of the identified structures. The identity of factor structures was shown among people of different nationalities, nationalities, people with different levels education and mental health. An important conclusion follows from this - since the structure of spaces is identical for different subjects, the factorization results obtained on one group of people can be used on another group of subjects.
Somewhat later, Bentler and Lavoie expanded the semantic space by adding factors such as “reality,” “density,” “orderliness,” and “ordinariness” to “strength,” “activity,” and “evaluation.”
Using materials from Russian vocabulary, a group of scientists identified the following factors: “evaluation”, “orderliness”, “complexity”, “activity”, “strength” and a specific factor - “comfort”.
Below we will consider several types of semantic differentials.
Nonverbal semantic differential
In addition to scales using antonyms, Osgood attempted to use graphic oppositions instead. The subjects were offered pairs of any geometric shapes type: black circle - white circle, up arrow - down arrow. After that, they were called various words, and they had to choose the figure from the proposed pair that, in their opinion, was more suitable for this word. For example, for the word “happy,” most participants in the experiment pointed to images of something colorful, sharp, and clear. This experiment showed high level versatility. Practical use nonverbal semantic differential can be found in studies of visual reasoning.
Partial semantic differentials
For some individual conceptual classes (private), when the differential methodology was built, the emergence of new factors specific to these conceptual classes was demonstrated. An example is the scaling of political terminology, as a result of which factors - “assessment”, “strength”, “activity” - which are usually independent of each other, merged together. They could be described as follows: “benevolent dynamism” - “malicious impotence.” Therefore, partial semantic differentials are characterized varying amounts factors. According to Osgood, there is an interaction between concepts and scales in making judgments.
Personal semantic differentials
Among private semantic spaces, there is a variety called personal differential, built on adjectives that describe different qualities, both others, surrounding people, and yourself. The procedure for constructing such a semantic differential is no different. A person is taken as the subject ( a real man or a movie character) and is assessed on a bipolar scale of opposing adjectives.
Each researcher can create his own scale, but it is hardly worth doing this. It is better to choose a scale from among standard scales that are original in the sense that they have their own proper name, are widely used and are included in the most commonly used scale system. They are also called original. Next, four discrete rating scales are considered: Likert, semantic differential, graphic rating and Stepel, as well as a constant sum scale and a ranking scale.
Likert scale based on choosing the degree of agreement or disagreement with some specific statement. In fact, one pole of this essentially bipolar ordinal scale is formulated, which is much simpler than naming both poles. The formulation of the statement may correspond to the ideal level of some parameter of the object. When characterizing the highest educational institution we can consider the following properties: qualified teaching staff, classroom equipment technical means, modernity and regularity of updating training courses, availability e-leming V educational technologies, level of culture, image and reputation, student population and many others. The wording of the statements could be as follows: the teaching staff of this university is very qualified; the university has a very high level of use of modern teaching aids; this university educates students seeking knowledge; graduates of this university are highly valued in the labor market.
When using a Likert scale, five gradations are usually considered. An example of using a Likert scale in a questionnaire is shown in Fig. 8.1. In other words, the questions are formulated in a Likert scale format. The respondent is asked to tick one of five boxes.
Rice. 8.1.
At the same time, she herself quantification In this case, the respondent is not required to do so, although more often points can be immediately given next to the names of the gradations. As can be seen from Fig. 8.1, the degree of agreement or disagreement with each statement made can have the following gradations: strongly disagree (1 point), disagree (2 points), neutral (3 points), agree (4 points), definitely agree (5 points). Here in brackets is the most commonly used option for digitizing the scale. 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 presupposes the presence of two polar semantic meanings (antonyms) or antonymic positions, between which there is an odd number of gradations. In this sense, the scale is bipolar. As a rule, seven gradations are considered. The middle position (middle 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 specified verbally (verbal). Examples of scales with two poles are as follows: “calming – 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 graphical arrangement of an odd number of gradations located between the extreme values “good - bad” is given.
Semantic differential method (from Greek. sematicos – denoting and lat. differentia difference) was proposed by the American psychologist Charles 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.
Can be considered as an analogue of the semantic differential scale. The rating scale is implemented in such a way that each property is associated with 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.
- Thus, when characterizing the exterior of certain car brands, they sometimes say that it is characterized by brutality. There are more simple examples– ergonomics and controllability, when it is difficult to meaningfully name the second pole.
B. P. Gromovik, A. D. Gasyuk,
L. A. Moroz, N. I. Chukhrai
Using the semantic differential in marketing research
Lviv State Medical University named after. Danil Galitsky
State University "Lviv Polytechnic"
IN modern conditions The need for marketing information is constantly growing, and marketing managers lack reliable, relevant and comprehensive data. To solve this problem, pharmaceutical enterprises must 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:
| internal reporting subsystem of a pharmaceutical enterprise, which makes it possible to track indicators reflecting sales levels, costs, volumes inventory, cash flow, data on receivables and payables, etc.; | |
| 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 special research into a specific marketing problem; | |
| analytical marketing subsystem, consisting of a statistical bank and a bank of mathematical models and covering advanced tools for analyzing data and problem situations. |
If external and internal information systematically accumulated in a marketing information system through market monitoring turns out to be insufficient, there is a need to conduct special studies of various marketing problems.
The marketing research process takes place in several stages (Fig. 1).
Rice. 1. Market research process
At the first stage, it is necessary to determine the subject of research and goals, which must be clearly defined and realistic.
The objectives of the study may be:
There are two types of marketing information collected during 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, marketing research, after processing and analyzing secondary information, proceeds to collecting primary data, which requires careful preparation. The plan for collecting information should primarily determine the research method. The most used research methods are presented in Fig. 2.
Rice. 2. Methods for collecting primary information
Observation is an analytical method with which the researcher studies the behavior of consumers, sales personnel; sometimes he acts as a participant in events (active observation).
A survey involves finding out people’s positions, 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 consumer behavior and his reaction to the design or advertising of a product.
If market research is insufficient, it is necessary to:
Most often used:
- trade panel (especially retail panel);
- consumer panel (end consumers or consumer organizations).
Experiment - a method with which you can study (find out) the reaction of the studied group of people to certain factors or their changes. The experiment aims to establish cause-and-effect 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 relevant 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 market 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-ended 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 answer options and the respondent chooses one of them. Forms closed questions may be different. The most common are alternative questions (assuming “yes” and “no” answers) and questions with selective answers. Quite often, researchers use various scales, in particular:
The stages of marketing research using semantic differential are presented in Fig. 3.
Rice. 3. Stages of marketing research using semantic differential
At the first stage, it is necessary to select 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. Next, the consumer characteristics of this product category that are most important for the target group of consumers under study are determined, and a system for assessing these characteristics is selected. After this, a questionnaire is developed to build a semantic differential. The next stage is a survey of consumer respondents, i.e., their construction of semantic differential curves, guided by the perception of the characteristics of the product under study, the basic competitor product and the hypothetical ideal product. Marketing research is completed by constructing average curves based on consumer opinions and analyzing each consumer characteristic of the products being studied.
As an example, we chose the “Magic of Herbs” shampoo as an object of marketing research, produced by the Nikolaev Pharmaceutical Factory and JV LLC “Magic of Herbs”. The comparison base was Elseve shampoo produced by the French company L’oreal.
These products were examined according to 10 consumer characteristics, which were assessed on a 10-point scale (table). Respondents rated each item on the questionnaire with a corresponding score for the “Magic of Herbs”, “Elseve” shampoo and the ideal shampoo that they would like to purchase.
Table. Semantic differential of consumer characteristics of “Magic of Herbs”, “Elseve” shampoos and ideal shampoo 
Based on the data obtained, average profiles of three curves were constructed, which reflect the average subjective perception of consumer characteristics of the products under study and the vision of the ideal shampoo.
Analyzing the curves (table), it should be noted that the studied shampoo “Magic of Herbs” satisfies target consumers according to the following characteristics: pleasant smell; effect of purity and silky shine; relatively well-known brand of product and the presence of natural ingredients; price (lower than Elseve shampoo).
At the same time, consumers are not entirely satisfied with the packaging of the Magic of Herbs shampoo, in particular, its design and convenience, as well as the lack of conditioner. Therefore, we can recommend that the manufacturer pay more attention to improving packaging and combining shampoo with other components (conditioner, keratides, etc.). Attention should be paid to the availability of sufficient quantities of shampoo in the retail network as a factor in the availability of its purchase.
Thus, the use of semantic differential in marketing research provides a thorough and visual differentiation of the characteristics of the compared products. In addition, it helps to identify the needs of different categories of consumers before choosing a product’s place 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. Marketing in Pharmacy: Pidr. for pharma universities and faculties / Ed. Z. M. Mnushko. - Kharkiv: Osnova, UkrFA, 1999.- P. 237–241.
- Starostina A. O. Marketing research. Practical aspect - K.; M.; SPb: View. house "Williams", 1998.- 262 p.