Rodrigo Uribe Bravo* & Carolina Martínez**

** Researcher. Business School. The University of Chile. Correo electrónico: cmartinezt@fen.uchile.cl

Abstract

This study examines children's consumption request in the context of a Latin American country. Concretely, this paper analyzes the amount and nature of gifts requested by Chilean children in their Christmas letters to Santa Claus in 2005. Results suggest, on the one hand, the presence of a small amount of gifts requested (as previous studies have stated), the gendered nature of them (in under-11 year-old kids) as well as a low presence of brands (as compared with figures reported by previous research). On the other hand, this study found a significant role played by age and gender in terms of type of gift and brand usage in the request.

Key words:

Christmas, children's request, brands, gender, age, Chile.

Resumen

Este estudio desarrolla un análisis de contenido de una muestra aleatoria de cartas escritas a Santa Claus por parte de un grupo de niños chilenos. El objetivo fue analizar la cantidad y naturaleza de las peticiones de consumo infantiles en términos del rol del género, edad, presencia de marca y estilo persuasivo usado. Los resultados muestran que las variables sexo y edad son de alta relevancia para explicar diferencias. En el caso de la primera, se detectó diferencias en qué regalos solicitaban y en la presencia de marcas en las peticiones. En el caso de la edad, se detectaron cambios en relación número de peticiones, el tipo de regalos solicitados, en el uso de las marcas y en el estilo de la petición realizada.

Palabras clave:

Résumé

L'étude développe une analyse du contenu d'un échantillon aléatoire de lettres écrites à Saint Nicolas par un groupe d'enfants chiliens, l'objectif étant l'analyse de la quantité et de la nature des demandes de consommation des enfants, suivant le rôle du genre, de l'âge, de la présence d'une marque et du style persuasif utilisé. Les résultats obtenus démontrent que les variables sexe et âge sont de grande importance pour expliquer les différences. Pour la première variable, des différences sont détectées concernant le genre de cadeau souhaité et la présence de marques dans la demande. Pour l'âge, des changements sont détectés par rapport au nombre de demandes, au genre de cadeau de mandé, à la marque utilisée et au style de demande réalisée.

Mots - clefs:

Noël, souhaits des enfants, marques, genre, âge, Chili.

Resumo

Este estudo desenvolve uma análise de conteúdo de uma amostra aleatória de cartas escritas ao Papai Noel por parte de um grupo de crianças chilenas. O objetivo foi analisar a quantidade e natureza das petições de consumo infantis em termos de papel do gênero, idade, presença de marca e estilo persuasivo usado. Os resultados mostram que as variáveis sexo e idade são de alta relevância para explicar diferenças. No caso da primeira, detectou-se diferenças em que presentes pediam e na presença de marcas nas petições. No caso da idade, detectaram-se mudanças em relação ao número de petições, o tipo de presentes solicitados, no uso das marcas e no estilo da petição realizada.

Palavras chave:

Décisions stratégiques, efficience, organisations publiques.

Introduction

Authors have used different notions to examine and interpret the behaviour of children (Gunter & Furnham, 1998; Valkenburg & Cantor, 2001). Some of them have emphasised the idea of "inter-generational influence" between parents and children (see Cotte & Wood, 2004 and Viswanathan et al., 2000) or the perspective of "family decision-making" in terms of the role of each member in relation to spending habits (Lee & Beatty, 2002; and Tinson et al., 2008). Nevertheless, the vast majority of the studies have assumed the viewpoint of "consumer socialisation", which emphasises how children's behaviour evolve as they mature and assume social roles (Roedder, 1999). For example, a recent study from Dotson and Hyatt (2005) provided an empirical explanation of how the consumer socialization process currently works with children. They detected five major influence factors: irrational social influence, importance of television, familial influence, shopping importance, and brand importance (Lueg et al., 2005).

Since children's income is limited, studies examining this issue have rarely focalized on kids' purchasing behaviour. With the exception of low-price products (those that little children can buy with their pocket money), an important part of research on this field has used kids' requests as a proxy measure of their consumer behaviour (Roedder, 1999). In this vein, Christmas has been considered a methodologically attractive moment of the year to examine children's request behaviour (particularly in the case of toys). Christmas time is an icon of contemporary commercial culture and a period in which kids may openly express their desires of possession to adults-parents or a symbolic figure such as Santa Claus in the case of under 8-9 year-old kids (Otnes et al., 1994a; Pesci et al. 1997; Okleshen et al, 2001).

The vast majority of the pieces of research examining children's request at Christmas have focused on the amount of things requested, the style of request and the presence of brands in it, stressing the role of gender as moderator of children's request behaviour (see for example, Otnes et al., 1994a). In this context, it is paradoxical that studies focusing on children have left the examination of the role of age as a relatively underexplored area, which is particularly clear in the case of the style of the request and the presence of brands in it. With the only exception of the seminal study of Caron and Ward (1975) and other from Isler et al. (1987) more than 20 years ago, there is no reported evidence about the particular changes in kids' request associated with the variable age.

Moreover, it is also important to note that all these studies examining children's request at Christmas have been carried out in developed countries such as the USA, New Zealand and European ones. Therefore, there is no available evidence from emerging economies -such as Latin- American ones-, which have shown a powerful economic growth and consumption patterns over the last decades (IMF, 2005).

This article seeks to continue and expand this line of research examining the nature of children's Christmas request behaviour in the context of a Latin American country -Chile. The focus of this analysis is the amount, type, style and brand presence in children's Christmas requests, which are separately explored as total prominence of them, as well as the differential role of children's gender and age (Roedder, 1999).

The analysis of Children's Request Behaviour

Since Christmas is a very significant moment of the year in terms of children's consumption behaviour (Okleshen et al., 2001), several studies have been carried out at that point in time to examine two main areas. The first one has evaluated the association between marketing actions and children's requests. Departing from the fact that TV advertising is the most important source of children's consumption ideas (Caron & Ward, 1975; Gunter et al., 2005), scholars have examined the association between Christmas request and advertising exposure. They have consistently detected that age, and sometimes gender, parental attitudes and mediation, and TV viewing patterns are able to explain significant differences in the influence that advertising has on children's requests (Pine & Nash, 2002; Robertson and Rossiter, 1974, 1976; Valkenburg and Buijzen, 2000).

The other relevant area of research on children consumption at Christmas-which is the focus of this study- is the analysis of the nature of children's requests. In particular, these studies have mainly explored four issues(see table 1): (1) how many products children ask for, (2) the type of gift requested, (3) the role played by brands in the request, and (4) the style of the request (Otnes et al., 1994b). Usually, searchers have examined these four dimensions in terms of the role of mediating variables such as children's gender and sometimes kids' age (Gunter & Furnham, 1998; Gunter et al., 2005; Valkenburg & Cantor, 2001; Ward, 1974).

Characteristics of Children's Request

The first relevant indicator of the nature of children's requests behaviour is the amount of gifts that kids ask for (Roedder, 1999). Regarding this issue, previous research has yielded dissimilar findings, with an amount of gift ranging from 3 to 7 per request (meaning letter). Findings have tended to be steady across gender, but not across age.

A seminal study conducted in the USA by Caron and Ward (1975) reported an average of 3.8 gifts per letter in a sample of American middle-class children, with ages between 9 and 12. They observed a significant decrease in the number of gifts requested as children grow older (they did not report gender differences), which is the only evidence available about the actual differences across age in kids' Christmas request. This finding provided support to other studies about children's request carried out in broader contexts. In this vein, a classical article published by Isler and colleagues (1987) showed that-in a common day-children from 3 to 4 years old asked for an average of 24.9 products a day (including food, toys, clothes, and other products). This figure decreased up to 13.3 items in the 5-to-6-year-old children and 10.4 products in those aged between 9 and11 years old.

In a subsequent study analysing letters to Santa -also carried out in the USA- Richardson and Simpson (1982) found a higher average of gifts mentioned in each note (5.3) examining a sample of kids of ages from 5 to 10 years old. This piece of research also noticed a small but not significant difference in the case of gender (an average of 5.6 in the case of boys vs. 5.0 in the case of girls). In these years, Bardbard and Parkman (1984) reported another study with pre-school children (from 3 to 5 years old). They found an average of 4 gifts per letter, detecting no differences by gender in this age group.

In the 90s, a classical investigation published by Otnes and colleagues (1994a) detected the highest number of gifts requested by children: 7.2 gifts per letter. These scholars could not establish significant differences in the number of gifts asked for by boys and girls (they did not report differences by age). In 1997, Pesci et al. reported an average of 4.6 gifts per letter without differences according the gender of children (they did not examine the variable age).

Evidence that is more recent comes from Pine and Nash (2002). These authors examined the Christmas letters of British and Swedish young children (5 to 6 year-olds). They found that these European kids asked for an average of about 5 gifts per letter. Simultaneous research conducted in Australia (O'Cass and Clarke, 2002), examined a sample of Christmas letters sent by children of different ages. These authors detected an average of 4.1 gifts per request, with no significant differences between boys and girls.

In this context, the current study investigates how these figures compare with a sample of letters written by Chilean kids. Then a first Research question is posited:

RQ1: How many Christmas gifts Chilean children request in each letter?

Additionally, in this paper we examine other two sub-research questions asking for the role of gender and, particularly, age as mediating variables in the number of gifts requested by Chilean children. It is important to bear in mind that previous research on Christmas request has not fully examined the latter aspect; so, this study seeks to provide more evidence that is robust, in this area:

RQ1a: Is there difference in the number of gifts requested by boys and girls?

Type of gift requested by children

Another relevant aspect of children's request refers to the type of gift. Exploring the postulates of several studies about gender and children, authors have examined how this variable is associated with the type of gift that children ask for. In this vein, Richardson and Simpson (1982) evaluated the incidence of masculine or feminine request. On all age groups examined, these authors reported that boys and girls tended to ask for different kinds of gifts: toys like dolls (and their accessories), domestics, educational art, or kitchen playsets in the case of the girls versus vehicles, sports, racing cars, or action figures in the case of boys. This finding was interpreted as an indication of a feminine socialization oriented towards the private, individual and domestic sphere as opposed to the masculine one, more directed towards the public space.

Revisiting Richardson and Simpson's study (1982), Bradbard and Parkman (1984) reported similar findings in their study with pre-school children. They used a methodology that divided kids' requests into 12 categories: dolls, domestic items, dollhouses, educational-art materials, musical items, spatial-temporal boys, sport equipment, furniture, vehicles/machine, clothing/jewerly and other. They observed that boys requested more items of sports equipment and vehicles/machines. Girls, instead, asked more frequently for dollhouses and accessories, clothing/jewerly and dolls. Nevertheless, they do not detected differences in other types of items such as educational art material, toy animals and musical items.

Based on the same principles, recent research conducted by Pesci and colleagues (1997) has directly classified kids' requests in terms of the gendered nature of them (in a three-point scale as masculine, feminine or neutral). Results showed that boys and girls tended to ask for those gifts traditionally associated to its gender. In fact, boys' requests were neutral toys (32%), masculine toys (65%), and feminine toys (only 3%). Instead, girls requested neutral toys (41%), masculine toys (13%), and feminine toys (46%). Although this classification is straightforward and intuitive, it omits that the gendertypicality of many toys is not very clear.

Thus, this piece of research examines the gendered nature of gifts requested by Chilean children in terms of a second research question:

RQ2: Do Chilean boys and girls ask for toys typically associated with the masculine and feminine sphere?

Subsequently, it is important to determine whether age is a relevant variable in the potential gendered nature of children's request. Then we explore the following subresearch question:

RQ2a: Are there are differences in terms of the gendered nature of gifts according to the variable age?

The presence of brands in the request

Regarding the first issue, studies have reported a range from 36 to 55% of requests incorporating brands, with some studies revealing an absence of influence of the variable gender. Otnes et al. (1994b) observed in the USA that over 50% of the gifts asked by children of different ages explicitly incorporated a brand, with no significant difference associated with the gender of children. In a more recent piece of research, Pine and Nash (2002) reported that European young children presented a lower level of brand presence: an average of 38% (this study did not examine the role of gender and age in the presence of branded requests). Finally, O'Cass and Clarke (2002)- examining a sample from Australia-found about 45% of the requests containing a particular brand with no differences by gender. It is interesting to note that there are no studies examining the role of age in terms of the presence of brands in children's Christmas request.

Based on earlier studies, this article examines the presence of brands in children's request in a sample of letters written by Chilean kids:

RQ3: What is the presence of branded request in the case of Chilean children?

Moreover, it is also explored the role of gender and age in the presence of brands in the gift request, being the latter a variable with no previous evidence reported in the literature:

RQ3a: Are there differences in the presence of brands in the request according to the variable gender?

RQ3b: Are there differences in the presence of brands in the request according to the variable age?

The other relevant issue related to the relationship between children's request and branding is the level of brand loyalty. Previous studies (Otnes et al., 1994a; O'Cass & Clarke, 2002) have proposed a continuum that runs from those petitions in which 100% of the request--no matter the number of gifts-is branded (brand-fixated) to requests in which the request includes two o more brands mentioned equally (pluralistic) (see Table 2).

]]>

Using this classification, Otnes and colleagues (1994a) reported that-among those letters including at least one branded gift-the most common pattern of brand loyalty was "brand pluralistic", followed by "brand dominant". That is to say, the analysis of brand loyalty showed that the most frequent situation was the inclusion of two o more brands in the request. Additionally, these authors did not observe significant gender differences in this pattern (they did not analyze the role played by age). On the other hand, the aforementioned study of O'Cass and Clarke (2002) detected a slightly different pattern. Although the most common form of brand loyalty was also "pluralistic" (same as Otnes and colleagues), the second most frequent was "singular brand". In other words, children from Australia showed a lower level of brand loyalty than American ones. Additionally, when examining potential gender differences, they detected no significant divergences between boys and girls (they did not report results by age).

In this area, this article explores the patterns of brand loyalty in a sample of Christmas letters written by Chilean children:

RQ4: What are the most frequent patterns of brand loyalty among Chilean children?

Children's brand loyalty is also explored in terms of the role of gender and age in two sub-research questions:

RQ4a: Are there differences in the most frequent patterns of brand loyalty according to the variable gender?

RQ4b: Are there differences in the most frequent patterns of brand loyalty according to the variable age?

The style of the request

The style of the request is another relevant way to examine children´s letter to Santa. In this vein, previous research has defined politeness and indirectness as the two main dimensions to examine this issue, which have only been examined as total and by gender. The former, defined as the group of rules about how the others have to be treated, has been repeatedly mentioned as one of the critical factors in the strategies of children persuasion (Blum-Kulka, 1997). In this issue, O'Cass and Clarke (2002) showed that 27.5% of the letters had courtesy elements (such as "how are you", "thanks", or "please"). Examining whether the variable gender could be a significant moderator of politeness, it was detected that girls used more frequently this element than boys did (32% versus 21%). This result was not consistent with the previous findings of Otnes and colleagues (1994b), which did not detected differences between boys and girls.

The use of indirect forms of request ("I would like...", "could you...") has been the other element used to examine children letters to Santa. The abovementioned study of O'Cass and Clarke (2002) detected that indirect request was massively used by Australian children (75% of the letters). Regarding the variable gender, Otnes et al. (1994b) detected an absence of significant differences in the use of this strategy, which was not supported by the study of O'Cass and Clarke (2002): girls tended to use indirect statements more frequently (81.9% versus 66.5% of the letters).

RQ5: How prominent is the use of polite and indirect requests in this sample?

This research question is explored in terms of the role of gender and age through two sub-research questions:

RQ5a: Are there differences in the presence of polite and indirect requests according to the variable gender?

RQ5b: Are there differences in the presence of polite and indirect requests according to the variable age?

Method

This study reports a quantitative content analysis of 500 letters addressed to Santa Claus sent through the Chilean postal service (Correos de Chile) during the four weeks prior to Christmas 2004. This technique of data analysis was chosen because it represents a replicable and systematic method to describe the manifest content of communication messages (Berelson, 1952; Riffe et al., 1998). Additionally, this is the technique used on previous studies examining the same issue (among others, Caron & Ward, 1975; Richardson & Simpson, 1982; Otnes et al.1994a; 1994b; O'Cass and Clarke, 2002; Pine & Nash, 2002), which provides a basic level of comparability for the results here yielded.

Regarding to the sampling procedure, it is important to note that Chilean Post allowed the researchers to select 10% of the letters to Santa received each day in order to be photocopied (and then returned)[1]. Then a sample of 10% (N = 500) of the total number of letters to Santa Claus received (Correos de Chile, 2004) was incorporated in the study. Subsequently, 122 letters were eliminated from further analysis for any or several of the following reasons (see Otnes et al, 1994a):

1. The senders did not include their name and/or age. The reason for this was that these elements allowed the researchers to determine author´s gender and age (86 letters).
]]>
2. The letter asked for generic gifts ("I want as many gifts as you can give me") or altruist wishes ("I wish peace in the world"), because they cannot be considered properly as consumption requests (18 letters).
3. Children did not write them ("I write to you on behalf of my 8 month old baby") or letters included adult gifts (35 letters).
4. Handwriting was illegible and then the content could not be understood (2 letters).

Thus, this study examined a final sample of 378 letters (see Table 3). They were written by a similar number of boys and girls (48.5% versus 51.5%), and an equivalent number of children in each age group (under 7, 8-10 and over 11 year old). Among those mentioning the address (85%), the sample belonged mainly to low income (87%) and urban zones of the country (98%) according to the classification of Chilean Market Research Association (Adimark, 2002).

]]>

The content of the letters was examined through of a group of codes compiled from earlier studies on children's request and adapted to the case of a Latin American country (see Annex). In order to ensure the reliability of the study, 7.5% of the sample (N = 30) was examined by a both coders (coders were undergraduate business students, with previous experience in content analysis). Then, the inter-coder agreement was computed according to Scot's Pi formula, showing a result over 0.8 across all the categories used in this study, which exceeds the minimum accepted to be included in the final data analysis (Riffe et al., 1998). Then, data was statistically examined (using T test and crosstabulation) to analyze the set of research questions (RQ) included in this paper.

Results

The amount of gifts

The first RQ asked for the number of gifts per request (Table 4). Results showed an average of 2.1 gifts per letter, with a statistical mode of 2. Further analyses by gender (RQ1a) detected a slight but not statistically significant difference between boys and girls. In fact, rhe results of the t-test revealed that there were no significant differences between the mean of gift requested by males and females as total and across all the age groups examined. On the other hand, the examination of the role of the variable "age" (RQ1b) indicated that younger children requested, in average, more gifts than older ones. In effect, 5-to-7 year old children requested significantly more gifts than 8-to-10 (t = 2,874, df = 291, p<0,005) and 11-to-13 kids (t = 3,353, df = 218,p<0,001). The same situation was detected in the comparison between the older groups where children between 11 and 13 years old requested less gifts than those between 8 and 10 (t = 2,452, df = 241, p<0,05).

Examining the level of dispersion. Ii is interesting to note that it tends to decrease as children mature. In fact, under-7-year-old children asked for a range between 1 to 5 gifts. The maximum number decreased up to 4 in the case of 8-10 kids and up to 3 in those children between 11 and 13 years old.

The type of gifts requested [T3]

The second area of analysis (RQ2) examined the genderspecific nature of requested gifts (Table 5). In order to do that, this study employed the classification of Blakemore and Centers (2005), which allows one to subdivide children requests into five subcategories: strongly masculine, moderately masculine, neutral, moderately feminine, and strongly feminine.

Findings show that girls and boys ask for those gifts associated to their gender (RQ2). Concretely, boys tended to massively ask for masculine gifts (63%) and girls did the same for feminine gifts (56%). Crosstabulations showed no differences between the distribution of the type of gifts according to the variable gender (x2 = 9.088, df = 6, p = 019). That is to say, in both cases the same tendency is detected.

A separate analysis by age (RQ2a) detected a significant influence of this variable in the request (x2 = 44.312, df = 7, p<0,001). Figures show that as children mature, a lower presence of gifts strongly and moderately associated with their gender is detected. In this vein, the oldest group depicted the highest presence of neutral requests and those moderately associated to the other gender (see Table 5).

The request for branded gifts

The percentage of branded requests (RQ3) reached 35.3% (0.9 brands per letter). Further analyses examined whether this figure was associated with the variables gender and age. Regarding the differences by gender (RQ3a), girls show a higher percentage of branded requests than boys did: 39.3% versus 31.8% (x2 = 29.088, df = 9, p<0,001). In terms on the average of brands, girls asked for 1.16 brands, and boys for 0.77, which depicts a statistical mean difference (t = 3,04, df = 376, 0,001).

The variable age (RQ3b) was also significant (x2 = 56.741, df = 9, p<0,001): results depicted a decrease in the incorporation of brands, as children grow older: 42 % in the youngest age group (under 7 years old), 35.6 % in the intermediate group (8 to 10 years-old) and 23.5 % in the oldest one (11 years-old and older). In terms of the average of brands per letter, the youngest group asked for 1.17 brands per letter versus the 8-to-10 group that requested 9.6 brands per letter (t = 2,13, df = 291, 0,005). Finally, the oldest groups asked for 0.41 brands per letter, which is also significantly different to the intermediate group (t = 3,946, df = 218, p = 0,001) and the youngest one (t = 4,938, df = 241, p = 0,001).

Patterns of brand loyalty

Brand loyalty was analyzed in terms of a classification proposed by Otnes et al. (1994a), which incorporates the different possible combinations of brand presence in the request (RQ4). The two most common styles of requests have in common the presence of only one brand, rather than two or more as reported by previous studies (Otnes et al., 1994a; O'Cass and Clarke, 2002). In fact, the patterns detected in this study revealed "brand fixated" (43%) is the most common request style in this sample (see Table 6). Those showing the "singularly branded" pattern follow it (38.7%). It is interesting to note that 60.3% of the users of this pattern asked for 3 or more gifts (over the average) and 29.1% of those requesting 1 or 2 gifts. That is, singularly branded request was also part of those letters with high number of requests (Table 6).

Further analyses examined the influence of children's gender and age in the pattern of brand loyalty. Regarding the former (RQ4a), crosstabulations revealed significant differences between boys and girls in terms of the style of brand request (x2 = 34.512, df = 9, p<0,001). In fact, girls mainly used both "brand-fixated" and "singularlybranded" requests (43.5%). In the case of boys, however, "brand- fixated" (44.7%) is the dominant pattern. "Singularly- branded" requests are in the second place, with lower presence (27.6%). Finally, in terms of the variable age (RQ4b), it was also significantly associated with the pattern of brand loyalty in the request (x2 = 47.139, df = 9, p<0,001). It is interesting to notice that the "brandfixated" request is considerably less found in the group of children over 11 years old, in which the "singularly-branded" request pattern becomes more prominent (Table 6).

The style of the request

The presence of courtesy elements is about 30% of the letters, with an average of 0.39 courtesy elements per letter. In terms of the role of gender, the average of polite elements was similar between boys and girls. Instead, the examination by age showed significant differences (X2 = 6.971, df = 5, p>0.05). Figures show that 12.6% of the letters written by 5-to-7-year-old children incorporated polite elements, which rose up to 33.7% in the case of 8-to-10 kids, and up to 46.4% in the oldest group. That is to say, as children mature they tend to use more frequently polite elements in the request (see Table 7).

In terms of the use of indirect request was in 23.6% of the letters. Again, the variable gender was not significant as source of differences in the use of this element. Nevertheless, the use of indirect requests was different according the variable age (X2 = 15.347, df = 5, p>0.001). Figures depicted that those children aged 5-7 had the lower use of indirect elements (13.9%). Then an increase of about 30% is observed in both the 8-to-10 and 11-to- 13 groups.

Conclusions

This study analyzed a Chilean sample of children's letters to Santa Claus in terms of the amount, gendered, branded nature and style of children's consumption request at Christmas. The analysis focused on the sex and age of children, being the later variable particularly relevant because of its underdevelopment in the field of both the branded character and style of the request.

A first important finding referred to the small amount of gifts requested in each letter analyzed: 2.1. Previous studies carried out in developed countries had detected an amount of requests ranging from 3 to 7 gifts per letter. Although there is no clear explanation for this difference, presumably this result can be related with a particularity of this sample: selected from an emerging economy (instead of developed country) and dominated by letters from poor backgrounds. Given this, it is possible to hypothesize that poverty imposes an environment dominated by certain limitations in the access to consumption, which would make a very long list of gifts irrelevant (Hill and Stephens, 1997).

In terms of the role played by children's gender and age in the number of gifts requested, findings showed that age was significant in terms of explaining differences among children. This finding reinforces the idea of a changing relationship with consumption as children grow older, which is a basic assumption of the theories of consumer socialization (Valkenburg & Cantor, 2001). Contrarily, gender socialization seems not to play any significant role in the quantity of the request, which is also consistent with previous evidence in the literature (Roedder, 1999; Gunter & Furnham, 1998).

A second examination of this study refers to the gendered nature of the request. Broadly speaking, the results reported in this paper support the idea that boys and girls tend to ask for those products associated with their gender role (about 60% of the requests). Nevertheless, it is interesting to notice a new finding. This tendency is less pronounced, as children get older. In fact, the 11-yearold group has the highest level of the neutral-gift request and the lowest levels of gendered request. It is important to be cautious with the interpretation of these findings, because they could be covering a potential change in the quality of gifts when children mature: from toys and games to other "non-gendered" things such as clothes and electronic devices (Fisher & Arnold, 1990).

Regarding the presence of brands, Chilean figures (35% of requests and 0.9 brands per letter) are in the lower range reported by previous studies (36% to 55%). Consistently, the analysis of brand loyalty shows dominant patterns of "brand-fixated" and "singularly branded" requests, which also demonstrate the lower presence of brands in comparison with other studies (dominated by "pluralistic" requests). Together, these results seem to demonstrate that brands would have lower quantitative prominence in the Chilean sample, which could also be related to the place in which this study was carried out: an emerging economy (in comparison to developedcountry kids).

Contrarily to previous evidence, this study detected significant differences between boys and girls in terms of the presence of brands. Girls also tended to include more "singularly-branded" requests than boys did. It could show a more flexible (or less loyal) incorporation of brands in the case of girls (Otnes et al. (1994b).

Finally, this study examined the style of request. On the one hand, the presence of polite elements was about 30%, which is the same level reported by previous literature. However, the presence of indirect request was lower than previous research: about 23% versus 75% in previous studies (O'Cass and Clarke, 2002). Tentative explanation for this difference with previous evidence is possible to find in the poor background of the sample, which has received an education of low quality (Perez-Arce, 1999).

In terms of the presence of gender, in both cases this variable did not show significant differences, which provided support to those results reported by O'Cass and Clarke (2002) but not to Otnes et al. (1994b) study. In the case of the role of age (a variable that literature had not previously examined), this study showed that the use of polite and indirect elements increased as children matured. Nevertheless, a difference was detected between both variables. Polite elements increased in the three age groups examined, but the use of indirect request only depicted a rise between under-7 and 8-to-10 children. This finding highlights the relevance of the variable age in the case of the style of consumer argumentation, which literature from other context -such as conflict resolution and education- had already stressed (see, for example, Bearison and Gass, 1979; Eisenbrg & Garvey, 1981).

All in all, these results show that gender and age are relevant factors to understand children's request. Across all the indicators, age produced significant differences in the sample examined. This result underlines the relevance of this factor to explain the differences in the consumers as consumer socialization literature has largely mentioned (Roedder, 1999). Similarly, gender was also associated with significant differences in many of the variables examined (the exceptions were the amount of gifts and the style of request). This finding also highlights the relevance of studying this variable as predictor of divergences in the development of children as consumers (Gunter and Furnham, 1999).

On the other hand, brands demonstrated a significant role in the request. This finding emphasises the relevance of brands from early childhood as previous literature about children branding has largely described it. (Achenreiner & Roedder, 2003; Lyndstrom & Seybold, 2003). This fact provides new evidence about the relevance (recognition and awareness) of brands in the context of a Latin American country, as well as the role played by age and gender in this process.

It is important to mention some limitations of this piece of research. First, the size of the sample stresses the tentative character of the results. In spite of representing a relatively large number of letters, it does not mean that results can be necessarily generalized. More research is needed in order to establish if the tendencies observed in this study are constant patterns, or they are affected by other elements such as the social, cultural or geographic characteristics of this sample.

Pie de página

[1] In Chile, children's letters to Santa are available in the main building of the Chilean Post for everyone who wants to read them and eventually send back a gift to the sender of the letter

References

Adimark. (2002). Mapa socioeconómico de Chile. Santiago: Adimark        [ Links ]

Bardbard, M.R. & Parkman, S.A. (1984). Gender differences in preschool children's toy. The Journal of genetic psychology, 145(2), 283-284.         [ Links ]

Bearison, D.J. & Gass, S.T. (1979). Hypothetical and practical reasoning: Children's persuasive appeals in different social context. Child Development, 50, 901-903.         [ Links ]

Berelson, B. (1952). Content analysis in communication research. New York: Free Press.         [ Links ]

Blakemore, J.E.O. & Centers, R. (2005). Characteristics of Boys' and Girls' Toys. Sex Roles, 59, 619-633.         [ Links ]

Dotson, M. & Hyatt, E. (2005). Major influence factors in children's consumer socialization. Journal of Consumer Marketing, 22(1), 35-42.         [ Links ]

Isler, L., Popper, E.T. & Ward, S. (1987). Children's purchase requests and parental responses: Results from a Diary Study. Journal of Advertising Research, 27(5), 28-39.         [ Links ]

Otnes, C., Kim, Y.C. & Kim, K. (1994a). All I Want for Christmas: An analysis of children's brand requests to Santa Claus. Journal of Popular Culture, 27(4), 183-193.         [ Links ]

Ward, S. (1974). Consumer socialization. Journal of Consumer Research, 1(1), 1-16.         [ Links ]