Nevertheless, researchers consider such restrictions as a rather radical step that may result in lack of consumer socialization process (Buijzen&Valkenburg, 2005; Nathanson, 2002; Opree, Buijzen, Reijmersdal&Valkenburg, 2014). If commercial content is reduced and a child stops receiving it, he also stops analyzing and criticizing it. That is why a child cannot become an emotional and cognitive well-educated consumer when he or she grows up. That is why it is important not just to reduce the frequency of brand interaction but also to teach children media literacy in families, kindergartens and schools. Statistically, at the age of 12 a child already has pocket money and basic consumer skills, though they do not have enough experience and cognitive maturity to judge about marketing manipulations and make considered rational decisions before purchasing. Children of this age tend to purchase fast and think impulsive in regard to brand preferences (Valkenburg & Cantor, 2001).

Children under 12 are not yet able to distinguish commercial content from non-commercial. What is more, they are critical towards themselves regarding what they own (Sirgy et al, 2012; Šramová, 2015). They tend to believe the advertising message they see and this makes their materialism level increase. Their purchase desire is at its highest stage and is easy to be influenced. Psychologically, teenagers feel defective because of not owing a product or brand advertised. This may proceed to a rapid decline of the level of psychological wealth as the consequence of the self-esteem reduction, changes in values, habits and attitude towards the society and culture surrounding (Fournier & Richins, 1991; Kasser, Ryan, Couchman & Sheldon, 2004, Šramová, 2014). Last but not least, it is important to take into consideration the fact that the ability to critically analyze marketing content is connected to the level of maturity, cognitive competence, curiosity, imagination, desire to study and self-education skills (Šramová, 2014).

As a result, there are some ways to generally reduce marketing effects on children's mentality by legislative limitations or teaching children consumer media literacy from their early age. Consumer media literacy can be identified as the process of development critical thinking and behavior, some skills and attitudes towards the whole process of purchasing: from the choice stage till purchase post-analysis and re-thinking stage (Šramová, 2014). So, this is a process when a child learns how to be a consumer. Since every educational process lasts the whole life, researchers and social-related group describe the aims of such educational process as teaching a child how to develop critical approach to marketing communications' perception of any type while upbringing, as well as obtaining ability to distinguish commercial from non-commercial offer, perceive advertising reality properly and competence to rational consumer decisions (Hradiská, Breèka & Vybíral, 2009; Šramová, 2014). One of the most important aims is to get to know how to select information and media sources instead of passive perception of content suggested by commercial sources.

There is one more problem concerning consumer media literacy teaching. In order to teach anything anyone, a teacher should be well-educated. Regarding the issue, a parent or a school teacher should himself possess consumer media literacy. Researchers claim that unfortunately, not every mature person is educated media literacy enough to teach younger generations. Especially, this can be referred to the states of post-communistic regime because in these societies advertising industry develops slower, so grown people are not well-educated enough regarding media literacy. They often do not even have rational consumer thinking experience and therefore, cannot teach it children (Schor, 2004; McNeal, 2007). This may be crucial limitation of the issue.

Then there comes a question of how to teach younger generation to analytical thinking in terms of commercialized content (Šramová, 2014).

First of all, parents or teachers should filter commercial information and the other one that can affect child's mentality even indirectly. By this unfair commercial information is meant: information that sounds ambiguously, inspires certain way of life and thinking, sales product by inculcating, information that stereotypes and judges behavior or possessions, establishes standards and promotes unhealthy way of life or non-moral principles and so forth.

Secondly, children trust parental argumentation, estimation and assessment (Buijzen, 2007). This is the reason why it is important for parents or teachers to preview and consume media commercial and non-commercial content along with children. Doing it together helps to more precisely and completely analyze the information received. By such discussions, a child can be taught how to perceive and understand media messages. When a child is educated enough to get, recognize and analyze marketing methods and intentions and formats of media influence on consumer, he or she will get to know how to be a consumer, how to develop consumer skills personally, how to make consumer decisions based on rational thinking rather than emotional (Opree, Buijzen, Reijmersdal & Valkenburg, 2014).



2. Empirical study

2.1 Operationalization

Loyalty can be measured by existence of several behavioral and perceptional patterns: having a sense of brand identity, having a previous positive relationship with the brand, a desire to return to the brand. In this research child and parental loyalty will be studied as two separate variables that reflect parental opinion about themselves and their children. So, further the variables “Child loyalty”, “Parental loyalty” will be analyzed. Furthermore, in the interviews and sociological survey, there was a question about the opinion of parents on the existence of correlation between parental loyalty and child's one. So, the dependence will be studied not just statistically, but also subjectively, so that the resulting values can be later compared with respondent's supposition.

Instant desire to buy means that a child is subjected to such behavior when he or she intends to possess an item if he / she faces commercial advertising or product placement activities. The answers for the respondents to reveal these values and their significance were as follows: whether a child or parent knows that this desire often happens and whether this factor is correlated with parental positive attitude towards the brand chosen by the child.

Materialism assumes presence of pleasure feelings, satisfaction, happiness received from shopping and product consumption, assessment of the quality of life of other people, depending on the product ownership or regular purchases. Moreover, in the survey, parents will answer the direct question whether they think their child is materialistic to any extent. Children will not be asked directly; the existence of this variable will be studied as an aggregate model with several factors.

The mode of communication in the family implies the predominance of communication aimed at liberal or conservative views on the information perception. This variable consists of several implicit factors for respondents, such as: openness in the relations between a child and a parent, taking part in common family decision making since an early age, the extent to which parents are ready to listen to child's opinion and give feedback, the extent to which children are free in their behavior and decisions, how children are controlled.

Socio-economical level of a family implies the variables dependent on the answers about parental education, income level, consumer behavior of parents and their materialistic ability.

2.2 Methods

Interview with children

During the study, seven interviews were conducted with five children aged 8-15 years. Gender differences do not significantly affect the perception of an advertising message. Therefore, neutral gender products were selected for the interview materials. Products that have gender predisposition were shown to a child of the relating gender. Age was chosen basing on the fact that by the age of 8 a child becomes an independent consumer: he or she has pocket money and experience in consumer behavior. In addition, by the age of 8, children have sufficient knowledge about the world, environment and brands, they already have basic values and preferences, they are aware of the existence of marketing influence and acquired basic media literacy, although the ability to resist advertising at this age is not yet well developed. In addition, by the age of 8, children can recognize the symbolism of a product and the brand image because they have already formed abstract thinking skills that contribute to brand's perception at conceptual level.

The upper limit is 15 years due to the fact that the interviewed children are still in the process of developing consumer skills and formation of materialistic values but not already formed.

The interview consisted of 5 blocks of questions (see Application E). Each block represents questions measuring the values that can be combined and create new variables, in order to reach the objectives of the interview. The questions were formed with the help of previous marketing research which this study is fractionally based on.

The tasks of the interview were to identify whether the level of child's materialism is related to the socio-economic status of the family; whether child's loyalty towards a brand is related to the family mode of communication; whether child's loyalty towards a brand is related to the parental attitude towards the brand; whether child's instant desire to purchase brand's goods is related to the family mode of communication; whether child's instant desire to purchase brand's goods is related the parental attitude towards the brand.

Interview with parents

The parents of the children participated in the interview were also interviewed. Six interviews with adults were conducted, on condition that it should be the parents of those children who have already been interviewed, because, in addition to testing hypotheses, it is important to check whether child's perception coincides with the parental one within one family. The further sociological survey was conducted on the sample of parents, that is why another task of this kind of researches was to measure the level of opinion coincidence between parents and children, in order to find similarities and discrepancies in child's and parental perception of variables researched.

The interview consisted of 2 blocks of questions (see Application F). The questions were as well formed with the help of previous marketing research which this study is fractionally based on.

The objectives of the interview were to identify whether parental relation to a brand is related to the family mode of communication and socioeconomic status of the family; whether parental relationship to the brand is related to the child's relationship to the brand; whether the desire to purchase brand's goods (instant or long term) by a parent has a correlation with the child's desire to purchase brand's goods (instant or long term).



Sociological survey

The results of a sociological survey are based on 207 valid answers received from 216 respondents. The survey was conducted on the Internet, in the form of a test with 32 questions: closed ones and questions with multiple choice. The sample consists of residents of Russia who are parents of children, on condition that their oldest child should be at least 2 years old because when having a 2-year-old child, a person has already developed parental patterns of behavior and also has experience managing child's desires and requests for purchase since at about the age of 2 a child begins to make the first requests.

There were 4 thematical blocks in the questionnaire (see Application G). They are: demographical chapter, part to reveal modus and socio-economical level of the family, the existence of child's materialism and both child's and parental loyalty and brand purchase desire.

The research is aimed at identifying correlation existence and its significance between the influence of family communication mode, socio-economic status of the family and the attitude towards the brand; correlation between child's attitude towards the brand and parental attitude towards the brand; correlation between child's desire to purchase brand's goods and parental desire to purchase brand's goods. The questions used in the current study intersect with the previous marketing studies where the same characteristics were separately examined.

marketing communications children brand parental

2.3 Results

The interviews were conducted before sociological survey not only to achieve its goals, but also to reveal new correlations or even variables.

The results of interviews with children and parents were comparable as children and parents within one family do perceive the main conceptions identically: among the conceptions there are family socio-economical level, communication modus, rights and restrictions, consumer behavior or decision-making processes. In other words, children and parents think the same conceptual framework within a family. The interviews revealed that the younger a child is, the less willingly he or she will talk with parents about the desired things from advertising. The older a child is, the less attention is paid to the source.

None of the respondents supposed that family relations can deteriorate due to the advertising impact on children. The results of this correlation were as well disproved in the survey.

During the interviews, some new variables for the survey were revealed. They contain direct questions about the correlations between child's attitude towards brands and parental one. In the questions of perception and attitude it is easier to directly ask respondents about their belief whether correlations can exist or not. Then, if the requests with direct correlations are affirmed by the majority of respondents, this correlation can be massively verified.

The results of the sociological survey with parents will be explained further step by step.

Descriptive statistics

For statistical analysis of the survey, SPSS 17.0 and data at significance level of 0.5 was used. The respondents were selected in the following way: there are 204 parents having one or more children, 57% of them are between the ages of 31 and 40 years old, 25% are between 41 and 55 years old, 13% are a group of 26-30 years old. 81% of respondents are women and 18% are men. 52% of respondents have one child, 47% of respondents have several children. The majority of responses (33%) is from parents with children aged from 6 to 12 years. So, predominantly, parents of the children of the target age group got into the sample. These socio-demographic characteristics of the sample are shown in Picture 1 and in Tables 2, 3, 4. The further tables and indicators can be found in Applications.

In the Tables 5, 6, 7 the distribution of frequencies and average priorities are shown, according to the socio-economical level of the family and the mode of communication prevailing in the family, based on one parent's opinion. Table 5 shows that the majority of respondents (85%) have a high social level - this includes the level of education, pleasure that is gathered from work and income, - and the average economic level (43%) - this includes the level of income per family and the aim of family purchase making. Distribution within the family communication mode was based on whether children are involved in the family purchasing decision process, whether pocket money is given to a child, whether a child has the freedom to choose and make consumer decisions, whether parents teach their children to openly express their opinion, whether they control a child strictly or not. The majority of respondents have a social mode of communication (45%), 39% of respondents have a conceptual mode, and these values are comparable in the sample. 14% of the answers were irrelevant because the respondents' children have not yet entered the consumer age and it would have been incorrect to assess their mode of parental communication with children.

In addition, in Tables 8-14 there is information about the frequency distribution of answer options related to the parent's opinion about the influence of different variables on the child's opinion. It was found that children of 57% of respondents take part in family purchasing decisions, their opinion influences the common family decision. At the same time, 31% answered that their children are too small to take part in the decision on family purchases.

Approximately equal number of respondents (91-92%) answered that if their child wants the product because of the advertising impact or because of their friends' influence, they would think about purchasing and did not give a definite answer in both cases.

Based on the recommendations of the previous studies, the interview addressed the issue of family relations deterioration due to the impact of advertising on children. Interview responses were negative: the prevailing point of view was that advertising does not affect the increase in discord within the family, but the quality of relations between children and parents does. This question was also investigated on a mass sample. The results of the interview were confirmed: 90% of parents answered that family relations cannot deteriorate due to the impact of advertising on the child, which confirms the findings of the interview with children and with parents (Table 15).

For the questions about correlation existence between preferences and brand selection by a parent and a child, 57% of respondents gave the answer that child's instant desire to purchase brand products is not related to parental preference for this brand. 82% of parents answered that children's loyalty is related to the loyalty of the parent. 88% of parents answered that parental attitude towards the brand is interconnected with child's attitude towards the brand. 55% of parents confirmed the reverse relationship that child's attitude towards the brand is associated with the parental attitude towards the brand.

All these results of the survey conducted on a mass sample confirms the results of the interview.

So, the purpose of the empirical part of research is to analyze the relationship between independent variables (which are gender, socioeconomic status of the family, family communication mode, giving pocket money to a child, sufficiency of time spent with children) and dependent ones (parent's opinion about the brand that the child wants due to the advertising impact or friends' influence, child's participation in family purchasing decisions, the desire to purchase brand's products by a child or a parent).

Correlations analysis. Before analyzing the data, it is necessary to look at the existence of correlations between the variables. In order to do this, SPSS 17.0 was used and the values of all responses from the survey were compared. Basing on the Pearson correlation coefficient, the correlations were analyzed. The coefficient indicates that the closer the Pearson coefficient is to 0, the less strong the relationship between the data is. And otherwise: the closer the coefficient is to 1 or -1, the stronger the connection between the data is. Having a positive coefficient, the correlation is positive. Having a negative coefficient, the correlation is negative. The significance of correlation is as well verified.

In the positive correlation, low value of one variable will correspond to low value of another variable. High value of one variable will correspond to high value of another variable. It was revealed that there are positive correlations between:

Variable-1	Variable-2	Pearson Correlation
Communication mode	child's age	0.348
Communication mode	social level of the family	0.161
Communication mode	family relationship worsening due to the advertising influence	0.160
Communication mode	child's loyalty towards brands	0.233
Child's instant desire to purchase due to parental positive attitude towards the brand	social level of the family	0.158
Child's instant desire to purchase due to parental positive attitude towards the brand	child's loyalty to brands	0.210
Child's instant desire to purchase due to parental positive attitude towards the brand	child's loyalty to a brand due to parental positive attitude towards the brand	0.184
Child's instant desire to purchase	child's participation in family purchasing decision making	0.167
Child's instant desire to purchase	child's loyalty to brands	0.533
Child's instant desire to purchase	child's loyalty to a brand due to parental positive attitude towards the brand	0.324
Child's loyalty to brands	child's participation in family purchasing decision making	0.168
Child's loyalty to brands	parental loyalty to brands	0.283
Child's loyalty to brands	influence of parental attitude towards a brand on child's attitude towards this brand	0.152
Child's loyalty to brands	giving pocket money to children	0.277
Child's loyalty to brands	personal child choice during purchasing	0.343
Child's loyalty to brands	parental opinion openness	0.298
Child's loyalty to brands	child's loyalty to a brand due to parental positive attitude towards the brand	0.525
Child's loyalty to brands	parental attitude towards a brand due to child's attitude towards the brand	0.152
Parental attitude towards a brand due to child's attitude towards the brand	Economical level of the family	0.140
Constant parental wish to purchase a product / brand that child wants	parental opinion on the brand that the child wants due to friends' influence	0.302
Constant parental wish to purchase a product / brand that child wants	parental opinion on the brand that the child wants due to the advertising influence	0.141
Parental opinion on the brand that the child wants due to friends' influence	parental opinion on the brand that the child wants due to the advertising influence	0.501

In case of negative connections, a low value of one variable will correspond to a high value of another variable. The high value of one variable will contribute to the low value of another variable. It was revealed that there is a negative relationship between:

Variable-1	Variable-2	Pearson Correlation
Communication mode	number of children in a family	-0.184
Communication mode	age of children	-0.444
Communication mode	age of a parent	-0.298
Communication mode	parental opinion on the brand that the child wants due to friends' influence	-0.146
Communication mode	parental opinion on the brand that the child wants due to the advertising influence	-0.080
Child's participation in family purchasing decision making	number of children in a family	-0.187
Child's participation in family purchasing decision making	age of a parent	-0.341
Child's materialism	Economical level of the family	-0.255
Child's instant desire to purchase	age of children	-0.266
Parental attitude towards a brand due to child's attitude towards the brand	social level of the family	-0.142
Parental attitude towards a brand due to child's attitude towards the brand	child's participation in family purchasing decision making	-0.409
Parental attitude towards a brand due to child's attitude towards the brand	personal child choice during purchasing	-0.250
Parental opinion on the brand that the child wants due to friends' influence	family relationship worsening due to the advertising influence	-0.648
Parental opinion on the brand that the child wants due to the advertising influence	family relationship worsening due to the advertising influence	-0.809
Child's loyalty to brands	age of a parent	-0.235
Child's loyalty to brands	sufficiency of time spent with children	-0.150
Child's loyalty to a brand due to parental positive attitude towards the brand	age of a parent	-0.204
Child's loyalty to a brand due to parental positive attitude towards the brand	sufficiency of time spent with children	-0.227
Child's attitude towards a brand due to parental attitude towards the brand	sufficiency of time spent with children	-0.155



Not all the correlations were analyzed in the current research due to the following reasons. First of all, not all of them make sense to the issues being researched. Secondly, as a result of the regression analysis, some dependent variables were excluded due to their insufficient level of significant correlation found in the regression model.

No correlation was identified due to the Pearson coefficient testing and no significance was found between the variables for testing H1: “Parental attitude towards brand is connected with the mode of family relations”. The existence of the correlation is disproved, H1 is disproved. The secondary check is shown in the Tables 16 and 17. This correlation will not be further considered in the study.

Also, no correlation was revealed between the factors that are considered in H4: “Child's desire of instant purchase is connected with the mode of family relations” and H6: “Child's desire of instant purchase is connected with the parental attitude towards brand”. The existence of a relationship is disproved, H4 and H6 are disproved. Secondary verification is shown in the Tables 18 and 19. This correlation will not be further considered in the study.

H2, H3, H5 about the correlation between child materialism and the socio-economic family status, child loyalty and family communication mode, child loyalty to the brand and parental attitude towards the brand will be checked further since the relationship between these variables is revealed.

Multiple regression. Then, the method of multiple linear regression was chosen since the aim of the analysis is to find out the relationship between several dependent variables and one independent variable, to identify the degree of influence of various factors on the target independent variable which is the variable that can be explained by the influence of dependent variables.

In the regression function, statistical significance of each dependent variable's influence is estimated. Then, it is possible to test hypothesis about the presence of statistical correlation between the variables.

At the same time, each dependent variable receives a coefficient that allows one to estimate the strength of its influence on the independent target variable and compare it with the strength of the other factors' influence.

The multiple regression model is a function (Ho, 2006):

Y = b₁õ₁ + b₂õ₂+ ... + b_nõ_n + a.

In this model “n” means the number of independent variables (regressors), denoted as “x”, “a” is a constant. Independent variables or regressors indicate the answers of respondents who had a certain gradation in the form of a Likert scale with 5 answer choices or fewer answers, as well as ordinal variables to which linear regression can be applied when interval scales are applied. These are the target variables that are affected by the independent ones. The example for target variables can be: child's attitude or parental opinion about the relationship or the degree of parental attitude. Predictors or dependent variables indicate variables that influence the independent one: gender, age, socio-economic status of the family, family communication mode, child's materialism, instant desire to purchase, loyalty, parental attitude towards brands.

The 95% confidence interval was used to determine the presence of correlation existence and significance level. With the statistical significance of the coefficients (p < 0.05), the hypotheses about the significance of the parameters were accepted, the hypotheses about the equality of the zero value of the parameters standing before the predictors were rejected, then the general model was assessed for its adequacy. The choice of predictors is crucial in identifying the optimal model explaining the independent variable. A step regression method was chosen. To find a better model, non-decisive factors were gradually removed from the model. The absence of multicollinearity was also checked to ensure that the independent variables do not merge with each other due to their strong correlation.

Besides testing the correlations in order to verify the remaining hypotheses, other possible correlations were investigated. Moreover, some correlations that might be of interest for future research, were as well identified. A total of 8 step-by-step analyzes of multiple linear regression were carried out in accordance with the number of independent variables.

First of all, the correlations to test the hypotheses of the study were examined.

H2 was confirmed: “Child's materialism is connected with the socio-economic status of the family”. The regression analysis revealed that the significant independent variable for the dependent variable “Child's materialism” is the “Economical level of the family”. The “Social family status” variable was excluded because its level of significance is insufficient for correlation. The “Family economic status” variable has a non-standard beta distribution coefficient of -0.330. The constant is 2.959, according to the model. Regression equation for predicting the importance of the economic status of the family is as follows:

Y = -0.330 * “Family economic level” + 2.959.

The variable “Family economic level” was coded as follows: 1 - Low, 2 - Middle, 3 - High. The calculation is as follows:

Y = -0.330 * 1 + 2.959, Y = 2.629;

Y = -0.330 * 2 2.959, Y = 2.299;

Y = -0.330 * 3 + 2.959, Y = 1.969.

Since there is a negative Pearson correlation between variables, a high value of one variable will correspond to a high value of another variable and vice versa. Then, the following dependence is obtained: the higher the Economical level of the family is, the higher is the level of child's materialism (Group of tables 1).

H3 was confirmed: “Child's loyalty towards the brand is connected with the mode of family relations”. The regression analysis revealed that the significant independent variable for the dependent variable “Child's loyalty” is “Family communication mode”. The “Family communication mode” variable has a non-standard beta distribution coefficient of 0.932. The constant is 2.959, according to the model. Regression equation for predicting the importance of family communication mode is as follows:

Y = 0.095 * “Family communication mode” + 0.932.

The variable “Family communication mode” was coded as follows: 1 - Social, 2 - Conceptual. Irrelevant answers were removed from the model. The calculation is as follows:

Y = 0.095 * 1 + 0.932, Y = 1.027;

Y = 0.095 * 2 + 0.932, Y = 1.122.

Then, the following dependence is obtained: the more conceptual the communication modus is, the higher is the level of child's loyalty (Group oftables 2).

H5 was confirmed: “Child's loyalty towards the brand is connected with the parental attitude towards brand”. By “parental attitude towards brand” several variables are implicated: “Parental loyalty to brands”, “Parental loyalty to a brand due to child's loyalty to the brand”, “Parental attitude towards a brand due to child's attitude towards the brand”. The last one was excluded because its level of significance is insufficient for correlation.

The regression analysis revealed that the significant independent variable for the dependent variable “Child's loyalty” are “Parental loyalty to a brand due to child's loyalty to the brand” and “Parental loyalty to brands”. The variables have a non-standard beta distribution coefficient of 0.386 and 0.342. The constant is 0.286, according to the model. Regression equation for predicting the importance of these variables is as follows:

Y = 0.386 * “Parental loyalty to a brand due to child's loyalty to the brand” + 0.286;

Y = 0.342 * “Parental loyalty to brands” + 0.286.

The variable “Parental loyalty to a brand due to child's loyalty to the brand” was coded as follows: 1 - Positive attitude, loyalty, 2 - No positive attitude, no loyalty. The calculation is as follows:

Y = 0.386 * 1 + 0.286, Y = 0.672;

Y = 0.386 * 2 + 0.286, Y = 1.058.

Then, the following dependence is obtained: the more loyal and positive to a brand a parent is, the more loyal is the child towards the brand (Group of tables 3).

The variable “Parental loyalty to brands” was coded as follows: 1 - loyalty existence, 2 - No loyalty. The calculation is as follows:

Y = 0.342 * 1 + 0.286, Y = 0.628;

Y = 0.342 * 2 + 0.286, Y = 0.970.

Then, the following dependence is obtained: the more loyal towards brands a parent is, the more loyal is the child towards brands (Group of tables 3). In order to verify the reciprocal correlation of these varieties, a new regression equation was conducted: the variable “Child's loyalty” has a non-standard beta distribution coefficient of 0.681. The constant is 0.428, according to the model:

Y = 0.681 * “Child's loyalty” + 0.428.

The variable “Child's loyalty” was coded as follows: 1 - Positive attitude, loyalty, 2 - No positive attitude, no loyalty. The calculation is as follows:

Y = 0. 681 * 1 + 0. 428, Y = 1.109;

Y = 0. 681 * 2 + 0. 428, Y = 1.790.



Then, the following dependence is obtained: the more loyal and positive to the brand a child is, the more loyal is the parent towards the brand (Group oftables 4).

Onwards will be verified correlations that were not considered as research hypotheses, although they may arouse interest for future research.

Using the Pearson correlation coefficient, it was revealed that there is some correlation between “Child's instant desire to purchase” and “Child's participation in family purchasing decision making”, “Child's loyalty to brands”. The “Child's participation in family purchasing decision making” variable was excluded because its level of significance is insufficient for correlation.

The regression analysis revealed that the significant independent variable for the dependent variable “Child's instant desire to purchase” is “Child's loyalty”. The “Child's loyalty” variable has a non-standard beta distribution coefficient of 0.650. The constant is 0.437, according to the model. Regression equation for predicting the importance of this variable is as follows:

Y = 0.650 * “Child's loyalty” + 0.437.

The variable “Child's loyalty” was coded as follows: 1 - Positive attitude, loyalty, 2 - No positive attitude, no loyalty. The calculation is as follows:

Y = 0.650 * 1 + 0.437, Y = 1.087;

Y = 0.650 * 2 + 0.437, Y = 1.737.

Then, the following dependence is obtained: the more the level of child's loyalty is, the less is child's instant desire to purchase (Group of tables 5).

Using the Pearson correlation coefficient, it was revealed that there is some correlation between “Parental attitude towards a brand due to child's attitude towards the brand” and “Parental opinion on the brand that the child wants due to friends' influence”.

The regression analysis revealed that the significant independent variable for the dependent variable “Parental attitude towards a brand due to child's attitude towards the brand” is “Economical level of the family”. This variable has a non-standard beta distribution coefficient of 0.097. The constant is 1.235, according to the model. Regression equation for predicting the importance of this variable is as follows:

Y = 0.097 * “Economical level of the family” + 1.235.

The variable “Economical level of the family” was coded as follows: 1 - Low, 2 - Middle, 3 - High. The calculation is as follows:

Y = 0.097 * 1 + 1.235, Y = 1.332;

Y = 0.097 * 2 + 1.235, Y = 1.429;

Y = 0.097 * 3 + 1.235, Y = 1.526.

Then, the following dependence is obtained: the more economical level of the family is, the better is parental attitude towards a brand due to child's attitude towards the brand (Group of tables 6).

Using the Pearson correlation coefficient, it was revealed that there is some correlation between “Parental opinion on the brand that the child wants due to the advertising influence” and “Parental opinion on the brand that the child wants due to friends' influence”.

The regression analysis revealed that the significant independent variable for the dependent variable “Parental opinion on the brand that the child wants due to the advertising influence” is “Parental opinion on the brand that the child wants due to friends' influence”. This variable has a non-standard beta distribution coefficient of 0.478. The constant is 1.099, according to the model. Regression equation for predicting the importance of this variable is as follows:

Y = 0.478 * “Parental opinion on the brand that the child wants due to friends' influence” + 1.099.

The variable “Parental opinion on the brand that the child wants due to friends' influence” was coded as follows: 1 - I will definitely buy it, 2 - I will think about purchasing, 3 - I will probably not buy it, 4 - I will definitely not buy it. The calculation is as follows:

Y = 0.478 * 1 + 1.099, Y = 1.577;

Y = 0.478 * 2 + 1.099, Y = 2.055;

Y = 0.478 * 3 + 1.099, Y = 2.533;

Y = 0.478 * 4 + 1.099, Y = 3.011.

Then, the following dependence is obtained: the more the parental desire to buy products / brands due to the friends' influence is, the less is the parental desire to buy products / brands due to the advertising influence (Group oftables 7).

Using the Pearson correlation coefficient, it was revealed that there is some correlation between “Child's instant desire to purchase” and “Age of children”. The variables of ages of 0-2, 2-5, 13-18, 19-25, 25+ years were excluded because its level of significance is insufficient for correlation and they do not relate to the audience of this study.

The regression analysis revealed that the significant independent variable for the dependent variable “Child's instant desire to purchase” is “Age of children”. This variable has a non-standard beta distribution coefficient of -0.538. The constant is 1.830, according to the model. Regression equation for predicting the importance of this variable is as follows:

Y = -0.538 * “Age of children” + 1.830.

The variable “Age of children” was coded as follows: 1 - Yes, the child is 6-12 years old, 2 - No, the child isn't 6-12 years old. The calculation is as follows:

Y = -0.538 * 1 + 1.830, Y = 1.292;

Y = -0.538 * 2 + 1.830, Y = 0.754.

Then, the following dependence is obtained: the closer child's age to 6-12 years old is, the higher is child's instant desire to purchase. (Group of tables 8).

Using the Pearson correlation coefficient, it was revealed that there is some correlation between “Child's attitude towards a brand due to parental attitude towards the brand” and “Social level of the family”, “Child's participation in family purchasing decision making”. The “Social level of the family” variable was excluded because its level of significance is insufficient for correlation.

The regression analysis revealed that the significant independent variable for the dependent variable “Child's attitude towards a brand due to parental attitude towards the brand” is “Child's participation in family purchasing decision making”. This variable has a non-standard beta distribution coefficient of -0.225. The constant is 1.840, according to the model. Regression equation for predicting the importance of this variable is as follows:

Y = -0.225 * “Child's participation in family purchasing decision making” + 1.840.

The variable “Child's participation in family purchasing decision making” was coded as follows: 1 - Yes, a child takes part in family purchasing decision making, 2 - No. The third variable “Children too small” was excluded because it is irrelevant. The calculation is as follows:

Y = -0.225 * 1 + 1.840, Y = 1.615;

Y = -0.225 * 2 + 1.840, Y = 1.390.

Then, the following dependence is obtained: the more likely child's participation in family purchasing decision making is, the less influence is child's attitude towards a brand due to parental one (Group of tables 9).



2.4 Discussion

Despite the number of researches conducted on the issue, it still remains unclear how to study perception and loyalty, these are processes complicated to explore in the longitude. So, in this research, loyalty is studied rather as an instant desire to follow brands in the future (to have loyalty). It does not prove whether loyalty will really exist in the future. Nevertheless, it is possible to measure real long-lasting loyalty only by observing respondents throughout their consumer life. It is also difficult to measure but can be included in future studies. So, the existence of correlations can be revealed by asking respondents on related questions. But physical, sociological and psychological aspects are difficult to combine, nevertheless they should be studied together in a common research.

2.5 Limitations and Future analysis

In this research, only the effects of commercial advertising were studied, the empirical part was built on a narrow sample of Russian respondents. Further studies can examine whether there exists the influence of the other types of marketing impact, as well as expand the sample and study impact characteristics in the countries with different culture and features from the Russian ones.

In addition, the study does not examine specific products or brands, its limitation is commercial communicative component. Therefore, it can be studied how children perceive certain categories of products or brands and how this is related to their parents' perception.

There was no difficulty in collecting data since this research is related only to marketing and communication research. Empirical methods are valid for the same field of study. However, consumer study field is not limited to communication. That is why, the results of previous studies can be further expanded with the help of this work basis but in other fields of study: psychology, social psychology, social and psychological development of a child, family psychology and sociology, physical cognitive features of perception of advertising impact and so on. For example, it is better to study child's mind and behavior in the correlation with child's biological cognitive processes explanation. Thus, since the research issues of preference and attitude and must obtain empirical data, it is impossible to penetrate into all the processes of the family relationship development. This proceeds to the next limitation: the short-term and long-term effects of advertising impact on the relationship between children and parents, are difficult to obtain due to the impossibility of analysis mass respondents' behavior constantly during the long period of time.

Moreover, this study does not represent verification of correlation existence between peer preferences in the context of brands, although this influence is significant, basing the results of previous studies. Social environment implies not just families but the other relatives as well. So, this research can be extended to the analysis of influence of the other relatives on child's perception and vice versa.

Further studies can also investigate which factors are correlated more with child's consumer behavior and parental one in the marketing context.



Conclusion

The study is aimed at exploring relationships between communicational models of parents and children within the family and the impact of commercial advertising targeted at children. The impact of advertising is being researched since 1950s and still stays actual. Children of even an early age (since the age of 0 years) become more and more involved into consumer decision making process and the process of developing consumer behavior skills. By the age of 5 children are already considered as independent consumers. From 8 to 14 years the main consumer patterns as materialistic values, preferences, judgements, negotiation and suggestion strategies, analysis and evaluation skills develop. The amount of money spent due to child's influence on their solvent parents reaches trillions of dollars worldwide and further increases annually (Solomon, 1999; Lindstrom, 2004). As far as all these processes' development dominates in childhood, parents play major role in the process of consumer socialization. This process is reciprocal: parents as well get re-socialized by their children in the issues of consumption and purchasing. Therefore, the connections between brand's impact, child-parental relationship and parental brand attitude were considered as key relations.

The research question raised in current study was whether commercial advertising targeted at children influence parental consumer decisions depending on the relations between them. This question was reviewed in details by means of the six hypotheses. Three of the hypotheses were disproved. Two of them were confirmed, one - partly confirmed. Five additional correlations were revealed while conducting the interviews, they were also were verified. As a result of the research, it was found that:

1. There is no correlation between parental attitude towards brand and the mode of family communication.

2. There is no correlation between child's desire of instant purchasing and the mode of family relations.

3. There is no correlation between child's desire of instant purchasing and parental attitude towards brand.

4. There is significant partial correlation between child's materialism level and the socio-economic status of the family. The variable of social level of a family does not play a significant role, it was rejected. The variable of economic level of a family showed significance in this correlation. So, the higher the economical level of a family is, the higher is the level of child's materialism.

5. There is significant correlation between child's loyalty towards the brand and the mode of family communication. The more conceptual the communication modus is, the higher is the level of child's loyalty towards the brand.

6. There is significant reciprocal correlation between child's loyalty towards the brand and parental attitude towards the brand. The more loyal and positive towards the brand a parent is, the more loyal is the child towards the brand. The more loyal and positive to the brand a child is, the more loyal is the parent towards the brand.

And the correlations that were not considered as hypotheses, but also included in the study:

7. The more the level of child's loyalty towards the brand is, the less is child's instant desire to purchase.

8. The higher economical level of the family is, the better is parental attitude towards a brand due to child's attitude towards the brand.

...