There has been a long standing interest in the economic understanding and financial literacy in children ( Moore & Stephens, 1975; Goodnow, 1998; Ashby, Schoon, & Webley, 2011 ). Many studies have found that parents play a key role in the financial socialisation of their children (Kim, La Taillade, & Kim, 2011; Pritchard & Myers, 1992; Webley & Nyhus, 2006) . Danes (1994) defined financial socialisation as “the process of acquiring and developing values, attitudes, standards, norms, knowledge, and behaviours that contribute to the financial viability and wellbeing of the individual” (p. 128).
Jorgensen and Salva (2010) suggested that children’s financial socialisation occurs through observations of their parent’s financial practices, positive and negative reinforcement received as a result of their own monetary behaviours, specific instructions received from parents, as well as role play and involvement in pocket money schemes (Alhabeeb, 1999; Bowen, 1996; Danes, 1994; Lachance & Choquette-Bernier, 2004) . Direct influences from parents may also include discussions about monetary practices, as well as keeping track of pocket money spending. This has been widely researched and parental involvement with financial education is suggested to result in an increased knowledge, as well as influencing the formation of attitudes towards money (Allen et al., 2007; Moore & Stephens, 1975; Moschis, 1985; Moschis, Prahasto, & Mitchell, 1986; Norvilitis & MacLean, 2010) .
Early studies looking at parents’ money education practices and children’s knowledge and use of money (Marshall & Magruder, 1960; Abramovitch, Freed- man, & Pliner 1991) found that children who received pocket money were more sophisticated with regards to financial decisions than those who did not. Lewis and Scott (2000) showed that that those who received pocket money throughout their childhoods were also economically more competent between the ages of 16 and 18 years.
Ashby, Schoon and Webley (2011) using a longitudinal study, concluded that 34 year olds’ saving practices were related to their socialisation experiences during adolescence, with a positive association being found between saving during adolescence and saving in adulthood. Barnet-Verzat and Wolff (2008) showed that more educated and wealthier parents give pocket money more frequently to their children. Such studies suggest a general consensus that wealthier families place more importance on regular pocket money giving and financial education than doing those from poorer backgrounds, showing socioeconomic status to be a reliable predictor of pocket money practices.
Furnham (1999) found that parents agree among themselves regarding wise allowance-giving systems for their children. The majority of parents (91%) were in favour of starting some weekly-based system by 6 years old, with the amount of money increasing linearly throughout childhood. There were clear age, but relatively few sex and class differences between the parents with respect to all the issues examined in the questionnaire. Furnham concluded that overall parents were consistent with regards to their attitudes towards the pocket money system which they had established, as well as how this economically socialised their child into knowledgeable and responsible attitudes to financial matters.
Further, research tends to reach agreement regarding the role of children’s age as a predictor of pocket money giving (Kerr & Cheadle, 1997; Furnham & Thomas, 1984; Feather 1991; Furnham, 1999, 2001 ). There is also much debate surrounding the role of sex differences between parent’s attitudes towards giving money (Furnham & Thomas, 1984; Furnham & Kirkcaldy, 2000) .
This study focuses on two things: the structure and content of (modern British) parent’s beliefs about pocket money and the predictors (demographic and ideological) of parent’s attitudes towards pocket money/allowance giving. Furnham (2008) reviewed a number of “popular books” aimed at giving parents advice as to how to make their children financially literate and responsible. These formed the basis of the questionnaires used in this study. This exploratory study predicted that attitudes would group into various factors concerned with how to educate and encourage money-related activities, as well as the setting of various rules. It was also predicted that parental income would be the major determinant of belief about economic socialisation.
There were 512 participants of whom 265 (52%) were male, and 228 (45%) female. They ranged in age from 18 to 67, with the mean age being 39 years. The dominant ethnicity of those taking part was European Caucasian (67.2%), with 12.3% being British Asian, 3.7% being African, and 11.1% documenting their ethnicity as “other”. With regards to income, 14.4% earn less than £15,000, 7.8% up to £22,000, 10.2% up to £30,000, 7.8% up to £40,000, 7.2% up to £50,000 and 40.8% earn more than £50,000, with 3.1% stating they were “unsure” about their income. Participant’s children ranged in age from 39 yrs to newborn. 431 participants had brothers, and 426 had sisters. The mean number of brothers was 1.22, and sisters was 1.34. The mean political orientation ranging from strongly right wing (1) to strongly left wing (7) was 3.74 (SD = 1.18). They also indicated how religious they were from Not at all (0) to Very (7) and the mean was 3.45 (SD = 1.93). Most of the participants would be considered to be middle class. They tended also to be very much moderates or centrists with respect to their political and religious beliefs.
1) Parental Rules and Regulations (Furnham, 2008) . This was a 43 item questionnaire that Furnham (2008) derived from a comprehensive review of books, papers and pamphlets on helping parents socialize their children into the economic world. The items can be seen in Table 1.
Departmental ethical permission was sought and received to undertake the study. The researchers explained to participants that the questionnaire aimed to explore opinions on children’s pocket money for a university research project. Once complete, participants returned their questionnaires to the researchers, who waited in the proximity. Participants were ensured that their answers would remain anonymous and that they could withdraw from participating at any time. Overall, participants took 10 to 15 minutes to complete the questionnaire. Having returned their completed questionnaires, participants were debriefed with regards to the aims of the study. Less than 5% did not complete the questionnaire.
aNote: Suppress absolute values <0.30. bParents attitude towards pocket money items; n = 43. Scale: Very Important 6 5 4 3 2 1 Unimportant.
3.1. Principal Component and Factor Analysis
Table 1 shows the mean and SD for each question. The statements that elicited most agreement about being important were 7 and 3, about giving clear explanations as to why a person cannot purchase every item they want, and 4, encouraging coin identification and change calculations. The two items which were felt to be least important were 5, which referred to starting pocket money systems as early as 3 - 4 years old, and 24, explaining taxation in family meetings.
Table 1 also shows the principal component analysis with varimax rotation for Scale 1 of parent’s attitude towards pocket money yielded five distinct factors. Together, the five observed factors accounted for 52 percent of the total
aNote: Suppress absolute values <0.30. bParents attitude towards pocket money items; n = 35. Scale: Agree 7 6 5 4 3 2 1 Disagree.
variance. The first factor labelled Discussion accounted for 33.2% of the variance, the second factor labelled Application accounted for 6.9% of the variance, the third factor labelled Home Practice accounted for 5.5% of the variance, the fourth factor labelled Responsibility accounted for 3.6% of the variance, and the fifth factor labelled Money Maturity accounted for 3.4% of the variance. In order to validate the appropriateness of the factor analysis for these data the Kaiser-Meyer-Oklin measure of sampling accuracy was considered, being .88, exceeding the recommended value of .60 (Kaiser, 1974) Overall items that loaded on factor 3 were judged most important and factor 5 least important.
Table 2 shows the mean and SDs for the second questionnaire. The participants agreed most strongly with items 25, 17, 24, 21, and 20 which all loaded on the first factor. Participants agreed least with items 9 and 28. The Table also shows that the principal component analysis with varimax rotation for Scale 2 of parent’s attitude towards pocket money yielded five distinct factors, accounting for 40% of the total variance. Factor one was labelled Examples accounting for 15.9% of the total variance, factor two labelled Accountability accounted for 8.0% of the variance, factor three labelled Freedom accounted for 6.5% of the variance, factor four labelled Organisation accounted for 5.0% of the variance, and factor five labelled Rules accounted for 4.3% of the variance. The Kaiser-Meyer- Oklin measure of sampling accuracy was .90, validating the appropriateness of the factor analysis for these data (Kaiser, 1974) .
3.2. Multiple Regression
A series of regressions were then computed with the scale scores as the criterion variable and five predictor variables: the participant’s age, gender, income, political and religious beliefs for each questionnaire.
First Questionnaire: The analysis was not significant for the first and fifth factor from the first analysis (see Table 1). The regression for the second factor was significant (F(5, 494) = 4.38, p < .001, AdjR2 = .04. Older (beta = .10, t = 2.21, p < .01) males (beta = .13, t = 2.57, p < .02) scored higher. The regression for the third factor was also significant (F(5,499) = 6.79, p < .001, AdjR2 = .05 indicated that male (beta = .17, t = 3.38, p < .001) less rich (beta = −.13, t = 2.62, p < .01), less religious (beta = −.11, t = 2.45, p < .01) and more right wing (beta = −.09, t = 2.04, p < .05) people agreed with this more.. The regression for the fourth factor was also significant ((F,5,495) = 10.18, p < .001, AdjR2 = .08). The only significant predictor was income (beta = −.30, t = .31, p < .001).
Second Questionnaire: The analysis for the five factors of the second scale showed three were significant. Factor two (F(5,493) = 3.71, p < .01, AdjR2 = .03) with Income being the only significant factor (Beta = −.17, t = 3.39, p < .001); Factor three (F(5.494) = 6.99, p < .001, AdjR2 = .06) with three significant predictors: sex (beta = .10, t = 2.20, p < .05), income (beta = −.14, 2.82, p < .01) and political orientation (beta = .10, t = 2.10, p < .05); and Factor 5 ((F5.493) = 2.97, p < .01, AdjR2 = .02) with sex being the only significant predictor (beta = .13, t = 2.56, p < .01).
Parent income (socio-economic status) was found to significantly predict the highest number offactors. This suggests that individuals with less income are more concerned about their children adopting efficient and effective money habits, through practicing these in the home, giving children responsibility to apply learnings, setting positive examples, ensuring accountability for the consequences of their monetary actions, giving their children freedom to make money related decisions, and enforcing rules surrounding pocket money. The significance of income as a predictor of attitudes towards pocket money is supported by past research ( Lewis & Scott, 2000; Barnet-Verzat & Wolff, 2008) .
However, the direction of the relationship discovered is interesting as it contradicts previous research such as that by Lewis and Scott (2003) , which suggests that financial activities are more common in families where parents have professional occupations. This may be explained by the difference in the two samples: This sample contained many more middle class people than in the smaller Lewis and Scott (2003) sample. In all 40.8% of participants earned over £50,000 (which is twice the average UK national wage) with the next most representative income group earning less than £15,000, with only 14.4% of participants fitting into this brackets
The finding in the present study that those earning less place more importance on monetary education may be due to the fact that those with lesser incomes have to enforce stricter management of their own money, and wish to pass on their knowledge with regards to effective spending and saving to their children. Cho, Gutter, Kim and Mauldin (2012) support this, concluding that those who earn higher incomes monitored their spending less frequently. It is interesting that those earning lower salaries also place more importance on giving their children monetary freedom than do those with higher incomes. However, giving children freedom having ensured home practice and setting examples may give children an opportunity to apply the learning’s for themselves in real life situations, and reinforce the effectiveness of such behaviours to them through reinforcement (Bar-On & Parker, 2000) . This reinforcement may be received in terms of praise from parents, or could be more direct, such as monetary benefits as a consequence of saving, for instance.
Gender differences were found to occur in terms of parent’s attitudes towards application of money learning’s and giving children freedom to manage their pocket money. Here, males placed more importance on giving children freedom to manage their money, while females placed more importance on practicing learning’s in the home, and setting rules surrounding pocket money. Such findings support Furnham and Kirkcaldy’s (2000) conclusion that parental gender was a significant predictor of beliefs and attitude toward allowances. Furnham and Kirkcaldy’s (2000) specific conclusions regarding males being more liberal with their pocket money giving and females being more in favour of applying rules were also supported, suggesting that these relationships are valid. Research by Warton and Goodnow (1991, 1995) also points to the reliability of this relationship, suggesting that females believe monetary rules should be clearly established and followed. Having said this, there is much past contrasting research, including Furnham (1999) concluding that there are few, if any sex differences in parent’s attitudes towards pocket money.
Ethnicity was also found to be a significant predictor of parent’s attitude toward pocket money, with non-Caucasian participants placing more importance on giving children responsibility (p < .001) and accountability (p < .001), where Caucasian participants placed more emphasis on setting rules surrounding pocket money (p < .001). This study therefore supports Scragg, Laugesen and Robinson’s (2003) conclusion that ethnicity has strong interactions with pocket money. However, further research is necessary in order to explore such relationships, and establish potential differences between Caucasian and non-Cau- casian parents. It is also important to highlight that the participant group was dominated by Caucasian participants (67.2%), and a higher number of non- Caucasian parents taking part in the study may altered the conclusions reached, and hence they may not be considered to be reliable. Therefore further research is required with a more representative sample to explore differences in parental attitudes towards pocket money between ethnicities. Further research may also consider cultural influences on pocket money practices, as we live in a diverse society, with mixed racial backgrounds, thus building on the differences identified between Caucasian and non-Caucasian parental practices.
Average age of a parent’s child was not found to predict participant’s attitude towards pocket money. This contrasts past research suggesting that pocket money giving relates to a child’s age, with many studies suggesting that this increases with age (Feather, 1991) . This may be due to the fact that the questionnaire did not address the amount of pocket money received by children as in past research (Feather, 1991; Furnham & Thomas, 1984) , with questions being more detailed regarding opinions surrounding aspects such as setting monetary examples to children and giving them freedom with their pocket money. However, Furnham’s (1999) study concluded that older children are more likely to be engaged in activities such as banking and saving. It may be that the age of a parent’s oldest child has the most significant impact on pocket money attitudes, setting a benchmark for practices for any younger siblings. Therefore it may have been more beneficial to consider this age in future research. Lewis and Scott (2003) for instance asked parents questions regarding finance related activities in relation to their eldest child, concluding that financial activities were more common in families with older children. Specific research in the area is lacking, but it may be that the pocket monetary system decided for the oldest child governs that which will be enforced for younger siblings. Further research is required in the area and only speculations can be made, but it may be that age of a parent’s oldest child predicts attitudes towards pocket money significantly, as in Lewis and Scott’s (2003) study.
Contrastingly, it may be that pocket money given to each individual child varies due to individual differences in the child. This would explain why age of child did not predict pocket monetary attitudes in the present study. For instance, if a child’s peers are receiving a certain amount of pocket money a parent may be likely to meet this so their child is able to afford to engage in the same activities as their peers (Sato, 2011) . Such a relationship may be responsible for confounding results regarding a child’s age and parental attitudes towards pocket money. Having said this, Furnham (1999) suggests that 62% of parents do not know how much pocket money their child’s peers receive, and few conclusions have been reached in the area. There may be cultural differences in this case, as Sato’s (2011) study is based in Korea, where there is a cultural system whereby peers exchange sweets and gifts, and thus similar amounts of pocket money are important so children are able to purchase such items in order to construct and maintain relationships. It would be interesting to investigate this relationship further, and control for it in future studies.
Such research could be applied in order to ensure all school children receive adequate economic socialisation regardless of differences in their parent’s attitudes. For instance, the present research points to the possibility that single parent families may have differing attitudes surrounding pocket money than nuclear families, due to the variance in views between males and females. Borgers, Dronkers and Van Praag (1996) provide empirical support for this inference, concluding that pocket money giving was significantly different in mother-headed families following divorce than in other family forms. However, it may be that divorce is acting as a confounding variable in this study and influences pocket money giving. Lont and Dronkers (2004) support this suggestion, concluding that pocket money rates are significantly higher in single-parent families than two-parent families.
This result is questioned when exploring Borgers, Dronkers and Van Praag’s (1996) study further in that pocket money in families headed by single mothers was also significantly different from that received by children in step families resulting from divorce. This undermines the suggestion that divorce is the underlying reason behind single-parent family systems differing from two-parent families. This finding instead indicates that the difference results from the establishment of a pocket money system by one gender. Therefore, it may be beneficial for schools to ensure that children from differing family set-ups are gaining sufficient economic socialisation, with no groups being disadvantaged. The present study also suggests that adequate economic socialization should be ensured for children of differing ethnicities and for those with parents earning differing salaries.
There is no doubt that nearly all parents are eager to ensure their children are financially literate and responsible. They do their by educating and socializing their children from an early age. Many are influenced in how their own parents taught them, though some adopt radically different practices. Not all are happy to acknowledge that children often observe and imitate the money-related behaviours of the parents as well as picking up good and bad habit from their friends.
This study also found that the parent’s social class and gender were related to their beliefs around financial education. We also found that the parents’ social class and gender were related to these beliefs. Many previous studies have also show wealth (rich vs. poor) and socio-economic (social class) differences in the way parents use their own money and socialize their children. Indeed there seems to be some inter-generational transfer on money beliefs and behaviours through this process. There is also evidence that with regard to socio-economic socialization fathers are more liberal and laissez-faire with their children than mothers. Future research would also benefit from accounting for more potential predictors of parent’s attitudes towards pocket money.
There are differences and contractions in books and papers that give advice to parents on socializing their children in general as well as specifically for financial literacy. This is one reason why this area of research is important as it attempts to show what particular socializing techniques have what precise consequences on children’s economic and financial beliefs and behaviors.