Segmentation of Women's Market Based on Personal Values and the Means-End Chain Model: a Framework For Advertising Strategy

Ved Prakash, Florida International University
ABSTRACT - The purpose of this paper is to examine the various bases for segmentation of women's market. Thereafter a conceptual framework is presented for the application of personal values (Rokeach Value Survey) and means-end chain model for identification of an advertising strategy. For hypothetical purposes Bartos'(1977, 1978) classification of women's market is used as the basis for segmentation. As a result of the hypothetical application of the means-end model for two products i.e. fashion clothing and automobiles,relevant benefits of the two products are identified. Suggestions for advertising strategy are offered; limitations of the study and directions for further research are provided.
[ to cite ]:
Ved Prakash (1986) ,"Segmentation of Women's Market Based on Personal Values and the Means-End Chain Model: a Framework For Advertising Strategy", in NA - Advances in Consumer Research Volume 13, eds. Richard J. Lutz, Provo, UT : Association for Consumer Research, Pages: 215-220.

Advances in Consumer Research Volume 13, 1986      Pages 215-220

SEGMENTATION OF WOMEN'S MARKET BASED ON PERSONAL VALUES AND THE MEANS-END CHAIN MODEL: A FRAMEWORK FOR ADVERTISING STRATEGY

Ved Prakash, Florida International University

ABSTRACT -

The purpose of this paper is to examine the various bases for segmentation of women's market. Thereafter a conceptual framework is presented for the application of personal values (Rokeach Value Survey) and means-end chain model for identification of an advertising strategy. For hypothetical purposes Bartos'(1977, 1978) classification of women's market is used as the basis for segmentation. As a result of the hypothetical application of the means-end model for two products i.e. fashion clothing and automobiles,relevant benefits of the two products are identified. Suggestions for advertising strategy are offered; limitations of the study and directions for further research are provided.

BACKGROUND

Marketers need to identify segmentation variables based on changing demographics, lifestyles, values and expectations of women (Joyce and Guiltinan 1978; Lazer and Smallwood 1977). An examination of literature on women's market shows that only in recent years have the consumer researchers begun to conduct detailed examination of the basis of segmentation. Since the 1970s increasing attention has been paid to women's career orientation. The literature on women's segment can be classified into three categories. The earlier studies only compared working women with non-working women. Examples of this category include Anderson (1972), McCall (1977), and Joyce and Guiltinan (1978). McCall (1977) compared working and nonworking women and found differences with respect to a number of buyer behavior activities. Joyce and Guiltinan (1978) conceptualized three categories of women i.e. housewives, non-professional working women and professional working women, and found significant differences with respect to retailing activities.

The second category of studies looked at the attitudinal (political) orientation of women. Douglas and Urban (1977) classified women on two lines i.e. traditionalist versus liberated and found significant differences between the U.S. and France. Reynolds, Crask and Wells (1977) divided women into Modern vs. Traditional lines and found Modern women to be more independent and pragmatic. Venkatesh (1980) divided women into three categories based on Arnott's (1972) Feminism Scale i.e. Feminists, Moderates and Traditionalists. He found significant differences in life style and magazine readership. The major problem with this type of segmentation is that it overlooks the employment status of women and motivation for work.

The third category of studies takes into account the employment status and motivation. Bartos (1977, 1978) went beyond the simple division of working wife vs. non-working wife. She presented a four-way classification: Stay at Home (29%), Plan to Work (20%), Just a Job (32%), and Career-oriented (19%). This classification was based on Yankelovich Monitor Survey at the time and Bartos found significant differences among the four groups with respect to travel, financial and shopping activities. Schaninger and Allen (1981) developed a three-way classification: Non-working Wife, Low Occupational Status and High Occupational Status. This classification is similar to Bartos' in that Low Occupational Status corresponds to the Just a Job type and High Occupational Status corresponds to Career-oriented women. Schaninger and Allen could predict differences among these groups with respect to food, beverage and alcohol consumption, make-up usage, clothing purchases, shopping behavior, media proneness, media wage, major and minor appliances. Nickols and Fox (1983) posited that income was a better predictor than occupational status for time-saving strategies. Stroeber and Weinberg (1977) and Weinberg and Winer (1983) found wife's employment status to be not related to family's time saving durables. Joag, Gentry and Hopper (1985) maintain that occupational status of the wife is insufficient for prediction of buyer behavior; consequently it is important to consider the work overload and motivation to work. These authors added another dimension i.e. homeward orientation vs. outward orientation to the classification suggested by Schaninger and Allen. Joag et. al. have not presented any empirical evidence for their scheme.

Research Gaps

There is no one best classification of Women's segment and many gaps remain in the area. First, none of the studies have examined various occupations of women i.e. sales, technical, managerial, academics and explored the differences in personal values. Secondly, except for the self-reports on attitudinal survey, there is no objective method of identifying segments. Thirdly, even for the segmentation schemes offered so far no effort has been made to explore the personal value structure on the lines suggested by Rokeach (1973); it is important to study the value structure in order to study the goals of employment. Finally, no attempt has yet been made to develop an advertising strategy based on personal values and means-end model (Gutman 1982).

WOMEN'S SEGMENT, VALUE STRUCTURE AND MEANS-END APPROACH: A CONCEPTUALIZATION

Personal values can be an important basis for segmentation; Personal Values can differ due to age, income, education, sex and social class (Rokeach 1973). Rokeach Value Survey (RVS) a popular methodology, consists of eighteen Terminal Values and eighteen Instrumental Values. Terminal Values e.g. Happiness, Sense of Accomplishment, Comfortable Life are enduring desired modes of existence. Instrumental Values e.g. being Capable, Forgiving, Broadminded are modes of behavior and are susceptible to change because of the socialization process. It is also possible that differences due to social class, income, education and self-concept would create differences in priorities placed on values by different women.

Just as personal values have antecedents they also have consequences e.g. impact on brand evaluation and choice criteria (Howard and Woodside 1984). Values also impact problem recognition, search for information, brand belief and ultimately brand preference. The impact of values can be visualized as follows:

FIGURE 1

Values impact choice criteria and are instrumental in determining benefit segmentation (Pitts and Woodside 1984) 1984). Empirically it has been shown that due to the differences in values, ethnic groups differ in product attribute perception of fashion clothing and automobile (Prakash 1984).

We can now draw on Gutman's (1982) means-end chain model. According to this model, ends are the valued states e.g. Comfortable Life, Exciting Life and products are the means to satisfy those values. In the context of women's market, it may be hypothesized that a study of the personal values can help in the development of an advertising campaign. We will first review the segments suggested by Bartos and hypothetically identify the value structure for each segment. This will be followed by a review of the means-end-chain model and its application for fashion clothing and cars.

Women's Segment and Value Structure [The discussion in this section is mainly based on Bartos (1977,1978,1981), Rokeach (1973) and Coleman (1983).]

The reasons for picking Bartos' (1977, 1978) Segmentation Scheme, i.e. stay at home, plan to work, just a Job and career-oriented are as follows. First, this scheme combines occupational status with motivation to work and consequently it can be tied to the value structure. Secondly, this scheme is intuitively appealing from female perspective is our conclusion based on in-depth focus group discussions. This scheme is also similar to Schaninger and Allen (1983) and Joag et al (1985) classification. The disadvantages of Bartos' classification are as follows. There is no objective method of measurement of the segments. The classification is based on attitudinal self-reports. Secondly, this scheme ignores income as a criteria; according to Nickols and Fox (1983) income is an important variable for time saving devices. Thirdly, the category of career-orientation is too broad; there could be further segmentation based on various types of careers. Bartos' scheme is selected because of simplicity, convenience and presentation of a hypothetical application of means-end model. The following identification of values is hypothetical based on focus group interviews. In an empirical study the respondents would have been asked to rank 18 terminal values and 18 instrumental values from the Rokeach Value Survey. Thereafter some type of cluster analysis would be performed to examine segmentation differences. Alternatively, one could look at the rank scores and perform analysis of variance to evaluate the differences.

The following is a hypothetical composite of the demographics o, each segment along with values extrapolated from the Rokeach Value Survey.

Stay at Home

Even though these women have no intention of working, they may at one time have been in the workforce, either before marriage or thereafter. They have no desire to change their life patterns, but some of them perceive it to be fairly limited. They are the oldest segment with more than half of them being over 50 years. This segment makes up approximately two-thirds of the total housewife segment.

This segment (based on age and social class) is likely to have less than high school education. It predominantly belongs to lower or working class (Coleman 1983). The lifestyle of this group is mainly traditional. Their lives are primarily bound by family and children. They have a somewhat lower self-concept and a defeatist attitude. They may sometimes long to be in the working woman's shoes in order to achieve self-fulfillment, may envy the latter's wardrobe, stimulating life and independence. On the other hand, they may perceive working women to be neglecting children and housework. They may describe themselves as kind, sociable, below the average in dominance, intelligence, broadmindedness and a sense of humor. Their buying style does not permit much advance planning. They are cautious and brand loyal. They have a mild interest in style and admit being persuasible. They are found to be far less involved in financial activities than the other segments. They are the heaviest viewers of daytime TV, second to "plan to work" in primetime viewing and listen to very little radio.

Based on the preceding description, the following Rokeach values were identified by the focus group to be important for this segment.

Terminal Values                   Instrumental Values

-Family Security                    -Loving

-Comfortable Life                 -Self-Controlled

-Self-Respect                       -Cheerful

-Salvation                             -Forgiving

-True Friendship                   -Obedient

-National Security                 -Helpful

-Inner Rarmony                     -Polite

Plan to Work

Some of the 'plan to work' housewives find that being a homemaker is personally rewarding. They are committed to their role as housewives but they also like to earn money. The plan of the working housewife carries some characteristics of the career woman. They make up the youngest segment with a median age of 32 mostly in the range of 2524 years of age. They are probably similar to Venkatesh's liberated group with few of the traditional/moderate values thrown in.

The educational profile of plan-to-work women is similar to just-a-job and career-oriented segments. They are slightly more likely to have graduated from college and to have completed high school education than just-a-job women. Assuming that most marriages take place into the same social class, because of the educational characteristics, it may be assumed that this group of women would mainly belong to middle and upper middle classes.

Women in this segment are athletic, cultured and energetic. They also allocate time for personal gratification and future career development. Their sense of the self is expressed in wanting to appear neat and attractive for self-pride. These women are a little more impulsive than "stay at homes". They are involved in financial activities similar to career women but to a lesser degree. They take an active role in shared decisions in the purchase of big ticket items such as cars. They are the heaviest viewers of primetime television and second heaviest readers of magazines (next to career women). This group is second behind career women in the purchase of convenience and time saving appliances.

Based on the above stated description from Bartos (1981), Venkatech (1981), and Coleman (1983), the following Rokeach Values were identified by the focus group for this segment.

Terminal Values                      Instrumental Values

-Comfortable Life                       -Broadminded

-Exciting Life                              -Responsible

-Social Recognition                     -Ambitious

-Family Security                          -Loving

-Self-Respect                             -Capable

-Sense of Accomplishment          -Imaginative

-Equality                                     -Intellectual

                                                  -Forgiving

Just a Job

This group makes up about 60% of the working women. They are not terribly involved with their jobs, but either enjoy the social aspect of self-fulfilLment of working, or work due to economic necessity. They usually finish high school and possibly some college, although to a lesser degree than plan to work or career women. The median age of this group is about 36, but a high ratio fall into ages below 25 and over 50, indicating a trend toward remaining at home during the child-rearing years which is confirmed by Yankelovich's claim that there is a movement from "plan to work" to the "just a job" category. Just a job women fall below career women in household income, but based on education/occupation factors, we may expect them to be found mainly in the working and middle classes. They correspond to low-occupational status category of Schaninger and Allen (1981).

This group appears to be split be been traditional/moderate and liberated viewpoints. The just a job woman may be described as trustworthy, kind, refined, stubborn, broadminded and affectionate. They are more concerned than career women with economy but are the least brand loyal and most experimental. They are responsive to new products and competitive promotional campaigns. They along with career women are likely to buy on impulse. They make an average use of financial services except investments where they are below average. They are independent decision makers in the purchase of large ticket items such as cars. Just a job women are a captive market for those using radio as an advertising medium.

Based on this information, the following Rokeach Values were identified by the focus group for this segment.

Terminable Values                       Instrumental Values

-Comfortable Life                          -Broadminded

-Exciting Life                                 -Cheerful

-Sense of Accomplishment            -Self-Controlled

-Social Recognition                       -Independent

-National Security                         -Responsible

-True Friendship                           -Polite

                                                    -Forgiving

                                                    -Honest

                                                     -Loving

Career-Oriented Women

They are professionally or technically trained women with a future in their career. Over 55% of them have attended or graduated from college. The median age of this segment is the same as that of "just a job" but they are more likely to be between 25 and 34 years of age. The implied viewpoint of this segment is likely to be liberated/moderate under the Venkatesh study. This group is the most affluent with 60% of those married to the most achieving men. It may be assumed that they are found in the middle, upper middle and lower upper classes (Coleman 1983). This group corresponds to high occupational category of Schaninger and Allen (1981) and Joag et al (1985).

Career women enjoy home surroundings a characteristic of the higher social classes and are brought up with the expectation of living in an orderly fashion. They usually purchase items that allow for convenience without loss of quality. The career women plan their shopping (more so on big ticket items) probably due to time constraints and are cautious and brand loyal. They do impulse shopping in order to reward themselves. They take a strong interest in grooming and style, mostly to maintain a social image. They are athletic and health conscious and are principal consumers of natural cereals.

Career women are the most active users of financial services and are the most likely to have purchased themselves the big ticket items such as a car. They are a prime target for travel both for business and pleasure. They are not heavy viewers of television. They are the heaviest readers of magazines and newspapers, the most active listeners of radio and heavy viewers of outdoor advertising. They may describe themselves as having a strong self-image, trustworthy, kind, refined, broadminded, efficient, intelligent, frank, self-assured, relaxed and creative.

Based on the above stated description, the following Rokeach Values were identified by the focus group for this segment.

Terminable Values                     Instrumental Values

-Comfortable Life                         -Imaginative

-Exciting Life                                -Intellectual

-Sense of Accomplishment           -Independent

-Self Respect                               -Ambitious

-Social Recognition                      -Broadminded

-Inner Harmony                           -Capable

-Mature Love                              -Honest

-Wisdom                                     -Loving

-World of Beauty                        -Self-Controlled

-Freedom                                    -Responsible

MEANS-END CHAIN MODEL

The means-end chain model establishes a framework to explain how products are the means to consumers' achievement of desired values. This model offers marketers a method of positioning products by associating product attributes (means) with benefits (consequences) that lead to the satisfaction of values (ends).

The means-end model developed by Gutman (1982) has four underlying assumptions:

1. values are dominant factors in the purchase pattern of consumers

2. consumers group products based on function to satisfy values

3. all actions have consequences (desirable or undesirable)

4. there is a link between consequences and product attributes

Gutman's model is developed on the premise that there are two major links in a consumer's cognitive map:

1. values are linked to consequences since consequences are given positive or negative valence depending on values

2. consequences are directly connected to product attributes since consumers will purchase the products they perceive will produce desirable benefits (consequences)

The linkages when combined have a common terminus i.e. values. Therefore, the basis of the model becomes the product attribute-consequence-value construct.

Based on the preceding model, we propose that the grouping of Rokeach Values within each of the four Bartos' segments would allow concentration on the centerpiece of the model i.e. consequences, in the development of advertising strategy for the target segments. Values, having been predetermined for a segment, become known variables with consequences and attributes being unknown. By utilizing techniques such as repertory grid (Kelly 1955), laddering (Gutman and Reynolds 1979), and product categorization (Gutman and Reynolds 1977, 1979), marketers can focus on the desirable consequences resulting from appropriate product attributes. The advertising message can emphasize the attributes that will help achieve valued end-states and ultimately lead to customer satisfaction.

ILLUSTRATION OF THE MEANS-END CHAIN MODEL

To illustrate the model incorporating the previously mentioned value groupings by segments for women's market, two products have been chosen: fashion clothing and automobiles. The set of matrices included at the end of this paper should be referenced to facilitate an understanding of the following analysis.

Fashion Clothing

To draw out the differences among the segments based on the grouping of values, consequences were held constant for the four segments. Values were taken from each segment based on a subjective association with the product class. Assuming that respondents were subjected to a method such as the repertory grid to ascertain level distinctions, a set of consequences similar to those in the attached matrices would be developed. A +1 or -1 is assigned to show the relationship between consequences and values.

In the case of fashion clothing, the following consequences were developed based on a focus group discussion: self-expressive, feel unique, easy care, affordable, personal comfort, practical/lifestyle and classic. These consequences were related to the respective values for each segment. A perusal of this set of matrices shows that for the stay at home segment, the highest consequence is that of easy care which is related to the values of comfortable life, self-respect and inner harmony. The other minor consequence is affordable which is related to the values of comfortable life and self-control. The "plan to work" segment is high in the consequences of feeling unique and being self-expressive; these consequences are tied to the values of an exciting life and self recognition. The "just a job" group also prefers the consequences of feeling unique and self-expressive. However, the strongest link is with the value of an exciting life. The "career" segment puts a high emphasis on enjoying a classic look, feeling unique and being self-expressive due to the values of being intellectual, imaginative and having an exciting life.

For this product, situations had to be considered as inputs into the system due to the different occasions for which one may purchase fashion clothing. Therefore, the second set of matrices (one for each segment) ties consequences to the situations. Some situations for which women may purchase fashion clothing are listed at the top of the second set of matrices; these are: leisure, sports, in the home, entertaining, appointments, party, evening and meeting. In the matrices the consequences are listed with the algebraic sign determined in the first set of matrices. To indicate the consequences that are relevant to specific situations a 1 is placed in the corresponding square The column for the situation of interest to the marketer is transferred to the top headings of the third set of matrices which connects the consequences (benefits) to the relevant product attributes. Assuming that a marketer were interested in gearing an advertising message to the "stay at home" segment for fashion clothing to be worn in the home situation, the product attributes to be stressed would be wash and wear, without detailed design and be flexible (i.e. can clean in, lounge in) connecting to the consequences (benefits) of easy care, personal comfort and practical lifestyle. An advertising strategy geared to the "plan to work" segment for fashion clothing to be purchased for a party situation should emphasize the attributes of separate garment, pieces to mix and match tying to the benefit of feeling unique; and also highLight new colors and wash and wear for consequences of self-expression and easy care respectively.

For the situation of entertaining, the "just a job" group is interested in receiving the same benefits as the "plan to work" group is for the party situation. However, the relevant benefits (consequences) have been ranked differently, exhibition a difference in priority. Therefore, the advertising strategy for fashion clothing aimed at the "just a job" segment would first emphasize new colors, then wash and wear, followed by separates for entertaining. Using the example of an evening out situation for the "career" category, a marketer would target this segment best by featuring product attributes of high styling, followed by current fad and new colors, tying to the benefits of classic. feeling trendy and self-expressive.

Automobiles

For the product category of automobiles it was necessary to construct two matrices for each of the four segments. The first matrix focuses on the values and consequences level to determine what the product class can do for the consumer. The second matrix (for each segment) focuses on the group level to uncover product attributes. The purchase of an automobile is not affected by situations but rather by economic status, social class and life cycle. Therefore, situations as inputs into the system are not used. To draw out differences among the segments based on grouping of values, consequences were held constant for the four segments. The following consequences were identified for cars with the help of focus group interviews: feeling safe, economical, affordable, feel stylish, personal comfort, and practical/lifestyle.

A look at the first set of matrices shows that the "stay at home" puts high importance on feeling safe, being economical and enjoying personal comfort consequences. These benefits relate to the values of family security. comfortable life and inner harmony. The "plan to work" group enjoys an automobile that is practical for their lifestyle, economical and safe relating to the values of comfortable life, family security and responsibility. The "just a job" segment would prefer the characteristic consequences of sporty, stylish and economical which ties to the values of independence, a sense of accomplishment and social recognition. In comparison the "career" woman ranks high on feeling sporty, stylish and enjoying personal options connecting to the values of intellectual, independence, sense of accomplishment and social recognition.

Transferring the highest ranked consequences to the second set of matrices for each segment, the relevant consequences are related to product attributes. Based on these matrices, the "stay at home" group would receive the greatest benefits from a medium sized car that has low maintenance, high mileage and is sturdy. These are the attributes to be stressed in an advertising message aimed at "stay at home" women. If the advertisers were interested in attracting "plan to work" women, they would gear advertising to emphasize compact yet plenty of room, sturdy, low maintenance but high gas mileage cars. To capture the "just a job" market, the marketers should advertise the qualities of compact, low maintenance, high gas mileage and a sleek design. The "career" oriented women would prefer luxury and sleekness in either a compact or a large car.

Implications for Advertising Strategy

Reynolds and Gutman (1984) show the importance of means-end model for promotional strategy. Basically the identification of values, situations and benefits helps in the development of ad campaign and slogans. For example, for the stay at home segment the important values are family security, comfortable life and self-respect. The plan to work woman values comfortable life, exciting life and social recognition. To the just a job segment important values are comfortable life, exciting life, and sense of accomplishment. Similar values are important to the career women as well. In product attributes stay at home enjoys personal comfort, feeling safe, being practical and economical in clothing. Plan to work desires to feel unique and self-expressive in fashion clothing and enjoying personal comfort in car. The "just a job" wants a sporty car a self-expressive fashion clothing in linking to an exciting life. The "career" woman desires a stylish sporty/luxury car and fashion clothing that is classic and trendy, which connects to the values of sense of accomplishment and exciting life.

This approach also helps decision on media, budget allocation and communication objectives. Through the process of laddering (i.e. hierarchical ordering of consequences and benefits) one can try to improve involvement for low involvement products (Gutman 1982). Finally this methodology helps in the development of marketing mix.

Here it is appropriate to compare the method described with some other methods of benefit segmentation (Myers 1976). In other methods of benefit segmentation multi-dimensional scaling or cluster analysis is performed on product attributes preferences. Those approaches are more quantitative and scientific. On the other hand the main characteristic of the means-end model is its qualitative approach giving freedom to management in decision making; the analysts in this approach moves slowly toward a hierarchical ranking of consequences in situations and establishing priorities. The means-end approach is simpler. The main disadvantage of this approach being relative lack of quantitative data. Of course, the major limitation of the present paper is that it is a conceptual study and no data is presented. This paper, however, provides guidelines for a future empirical study and we are in the process of implementation and data collection. A set of matrices is presented below as an illustration of the technique.

FIGURE

FIGURE

REFERENCES

Anderson, Beverlee B. (1972) "Working Women vs. Non-Working Women: A Comparison of Shopping Behavior" in Borris W. Becker and Helmut Becker (eds.) 1972 Combined Proceedings, American Marketing, Association, Chicago, 355-357.

Arnott, Catherine C. (1972) "Husband's Attitudes and Wives Commitment to Employment", Journal of Marriage and Family, 34, 673-81.

Bartos, Rena (1977) "The Moving Target: The Impact of Women's Employment on Consumer Behavior", Journal of Marketing, July, pp. 31-37.

Bartos, Rena (1978) "What Every Marketer Should Know About Women", Harvard Business Review, pp. 73-85.

Bartos, Rena (1981) "Beyond the Cookie Cutters", Marketing and Media Decisions, pp. 54-59.

Coleman, Richard P. (1983) "The Continuing Significance of Social Class to Marketing", Journal of Consumer Research, Vol. 10, pp. 265-278.

Douglas, Susan P. and Urban, Christine D. (1977) "Life Style Analysis to Profile Women in International Markets", Journal of Marketing, July 1977, pp. 46-54.

Ferber, Marianne A. and Birnbaum, Bonnie (1980) "One Job or Two Jobs: The Implications for Young Wives", Journal of Consumer Research, Vol. 7, pp. 263-271.

Ferre, Nyra Marx (1976) "Working Class Jobs: Housework and Paid Work as Sources of Satisfaction", Social Problems, Vol. 23, pp. 431-441.

Gutman, Jonathan (1982) "A Means-End Chain Model Based on Consumer Categorization Processes", Journal of Marketing, pp. 60-72.

Hafstrom, Jeanne L. and Dunsing, Marilyn M. (1978) "Socioeconomic and Social Psychological Influences on Reasons Wives Work", Journal of Consumer Research, Vol. 5, pp. 169-175.

Howard, John A. and Woodside, Arch G. (1984) "Personal Values Affecting Consumer Psychology", in Robert E. Pitts and Arch G. Woodside (eds.) Personal Values and Consumer Psychology, D.C.Heath Company, Lexington, Mass., pp. 3-12.

Joag, Shreekant G., Gentry, James W. and Mohler,JoAnne (1985) "Explaining Differences in Consumption by Working and Non-Working Wives", in Elizabeth Hirschman and Morris Holbrook (eds.) Advances in Consumer Research, Vol. 12, Association for Consumer Research, Provo: Utah, pp. 582-585.

(For a complete list of References, please write to the Author.)

----------------------------------------