Factors Affecting Purchase Frequency Of Clothes Among UK Students: A Quantitative Essay.

Effect of belief in climate change on purchase frequency of clothes

The study utilized a quantitative approach. Data was collected from a sample of 202 students from selected learning institution in the United Kingdom. The researchers used a structured questionnaire to elicit response from the participants with the aim of establishing factors that affect their purchase frequency of clothes.

To test for the effect of belief in climate change on the purchase frequency of clothes, a mean value of the multiple responses was computed to form a new variable “Avebelief” which was then used in subsequent analysis. 

A scale reliability analysis on the belief in climate change was conducted to obtain an acceptable value of Cronbach’s Alpha of 0.821. The exercise resulted in five scale items listed below;

• Climate change is just a natural fluctuation in earth’s temperatures
• Human activities have no significant impact on global temperatures
• The evidence for climate change is unreliable
• There is too much conflicting evidence about climate change to know whether it is actually happening
• It is too early to say whether climate change is really a problem

Report and interpret the results of analyses conducted to test the reliability of your measure(s) used in your survey.

The descriptive statistics is displayed in tabular format below;

From the table above, majority of the participants were 19 years of age (N=71,) represented by 30.9 percent.

The table above, shows the gender distribution of the research correspondents. The outcome indicate that the majority of the participants were female represented by 45.4 percent (N=104) whereas the males were represented by 37.6 percent (N=86). This distribution is further illustrated by figure 1 below;

Figure 1: Histogram illustrating the distribution of the gender of participants

From table 3 above, majority of the respondents indicated that there region of residence was in the United Kingdom (N=229) with only one person indicating a region of residence other than in the United Kingdom.

Table 4 above shows the education level of the participants. The majority of the correspondents indicated that they were students represented by 90.9 percent (N=221) whereas only 9 percent said that they were not students (N=9).

From table 5 of the nationality of the correspondents, approximately 50 percent indicated that they were of British origin (N=111). The least count of only one participant was spread across various nationalities which encompassed the following; UK, United States/ France, Welsh, werrfwer, White, Kazakh, Lithuanian, Mixed British, Portuguese, Russian, Spain, Spanish, Brazilian, British Asian, British Citizen, Bulgarian, China, ESPAÑAAA, Filipino, French/British, Greek/British, Hungarian, and Irish. The Italian, Ploish, and white British were the second dominant nationality (N=4).

The source of income of the participants is displayed in table 6 above. Majority of the participants nearly 50 percent depends on their parents for income (N=99), with 21.8 percent obtaining their income from part-time job (N=44). Those students depending on their student loan constituted 20.8 percent of the study population (N=42). Conversely, 6.4 percent (N=4) depend on the money in their saving accounts with 2 percent indicating other (N=4).

Table 7 displays the distribution of the amount of income that the research participants receive for a particular term. The majority of them fall within the low income category (N=128), whereas those in the middle-income range constituted 27.2 percent (N=55). The least margin constituted high-income participants (N=19).

Demographics of the research participants

Tests of Group Difference Hypotheses – Restate your hypotheses pertaining to tests of group differences. For each hypothesis, report the statistical analysis you used to test that hypothesis and why along with the results of that test including corresponding effect size indicators.  Provide a conceptual interpretation of the results regarding whether or not your results support your stated hypothesis. 

The testable hypothesis for this particular analysis is as listed below;

Null Hypothesis: There are no differences between genders in frequency of purchasing clothes.

Alternative Hypothesis: There are differences between genders in frequency of purchasing clothes.

A one-way ANOVA is used to test the validity of the above hypothesis and the outcome is as tabulated herein.

From the table above, the Levene’s test of homogeneity of variance is satisfied therefore accepting the null hypothesis on the equality of variance, F (2,211) = 0.342, p =0.711.

 From the ANOVA table above, the outcome is significant implying that the null hypothesis is rejected. Therefore, it can be concluded that, there are differences between genders in frequency of purchasing clothes, F (2,211) = 2.976, p =0.05.

To test the nature of interaction between gender and income group as an additional independent factor in explaining the frequency of purchasing new clothes in the United Kingdom, a two way ANOVA was performed.  This approach was selected since there is 3 values of gender and 3 values of moderating income variable hence feasible for performing a 3 by 3 two-way ANOVA. The outcome of the analysis is as presented below;

The table above shows the between-subject factors. The first factor is gender which has three levels, that is male (N=75), Female (N=90), and if not stated, please specify (N=37). The second factor income which also has three elves, that is low income (N=128), Middle income (N=55), High income (N=19).

From table 11 above, it is evident that on average, a higher purchasing frequency was recorded among the high income individuals (M=3.16, SD = 1.50), compared to the middle income persons (M=2.25, SD =0.985), and in the low income level (M=1.80, SD =0.732). Table 11 also shows that overall clothes purchasing frequency for the females and those that did not indicate their gender, was relatively similar (M =2.14 and M=2.27). Nevertheless, the purchasing frequency for the male participants was considerably low (M =1.84, SD = 1.027).

The Levene’s test output is shown in table 10 above showing that the assumption of equality of variances has been violated, therefore rejecting the null hypothesis on equality of variance, F (8, 193) = 4.995, p = 0.00. The ANOVA output is illustrated by the table below;

From table 11 above, the there is a positive significant interaction with a small effect  between the impact of  gender and income group in explaining the frequency of purchasing new clothes in the United Kingdom, F (4, 193) = 0.657, p = 0.623, ηp^2 = .013.

Consequently, the effect of gender on the purchasing frequency of the participants was non-significant (F (2,193) = 1.286, p > .05, ηp^2 = .013).

Source of income among research participants

Additionally, the large effect of income on the purchasing frequency of the participants was significant (F (2,193) = 17.920, p =0.00, ηp^2 = .157).

Figure 2: plot of estimated marginal means for the frequency of purchasing clothes in the United Kingdom

From figure 2 above, it is clear that the three gender groups performed relatively the same in the middle income category in influencing the purchasing of clothes in the United Kingdom. Nevertheless, this difference was significant in both the high income and the low income categories.

1. Tests of Relational Hypotheses –

The relational hypothesis is as listed below; 

Null Hypothesis: There is not a relationship between one's beliefs in climate change and frequency to purchase clothes.

Alternative Hypothesis: There is a relationship between one's beliefs in climate change and frequency to purchase clothes.

Before testing the relational hypothesis above, the reliability of the scale was computed by calculating the value of the Cronbach’s Alpha. The outcome is as shown in the table below;

Since the value of the Cronbach’s Alpha is below 0.50, the items selected for the scale are considered inappropriate and should therefore be selected again. To obtain an acceptable range of above 0.70, two items were deleted from the scale to achieve the below reliability statistics;

Table 15 shows an acceptable value of the Cronbach’s Alpha which is 0.821.

The items within the new scale are as depicted in table 16 showing that a total of 5 entries are utilized.

The items in table 16 above are transformed to form “Avebelief” which is the average of the scores of the 5 respective items. The outcome of the correlation analysis is shown below;

From table 17, significance value of the Pearson correlation is above the 0.05 threshold, therefore, the null hypothesis is accepted. This implies that, there is non-significant positive relationship between one's beliefs in climate change and frequency to purchase clothes, r (198) = 0.078.

To check for the effect of income as a mediation variable, a linear regression analysis was conducted. The outcome of the analysis is as presented below;

From table 18 above, there was a statistically significant association between belief in climate change, income, and  frequency of purchase of clothes in that the two predictor variables significantly explained purchase frequency, β = 0.881, t (202) =  3.684, p < .05.  

Income and belief in climate change also provided a significant explanation of the variability in purchase frequency of clothes, R^2 = .174, F (2, 195) =20.48, p = 0.00. Conversely, 17.4 percent of the variability in the frequency of purchase of clothes can be explained by the two predictor variables. 

To check on the effect of income and gender as a control variable, a multiple linear regression is conducted. The outcome is presented below;

The outcome in table 19 shows the contribution of each of the control variables in predicting the dependent variable which is the frequency of purchase of clothes. Income significantly predicted purchase frequency, β = 0.609, t (202) = 6.39, p < .05. The same observation was made for the gender variable, β = 0.250, t (202) = 2.861, p < .05.

The study faces one major limitation. The fact that the researchers collected data from participants from one institution affects the generalizability of the study findings. It, therefore, requires that prospective investigations should consider a wider geographical setting to boost the accuracy and applicability of the outcome of the general population.