Analysis Of Sales Performance Of A Small Health Food Shop On The Sunshine Coast

Research objectives

The report is designed to analyze the small health food shop on the Sunshine Coast performance, though put the year. The main focus will be on the average sales, gross sales totals profit and net sales of this business as they are the main measure of business progress. First, the coding of data was checked to ensure that they are appropriate. Notably, the month of the year, the season of the year, and the weekday variables were changed in the SPSS from ordinal to nominal. This is because, by Leech, et al., (2014) nominal variable are those that do their rank do not matter whereas the ordinal variables the ranks matter. For instance, the days of the year were left to be on an ordinal scale, and the month of the year, the season of the year and weekday were changed to nominal variables. The change was due to the fact that these variable ranks do not matter.

Statistical analysis, both descriptive and inferential statistics were performed to determine the characteristics, measure of central tendency and dispersion and to test the hypothesis. The research exclusively used business statistical approach to determine the state of the business.

The research was designed to achieve the following objectives.

The report was designed to achieve the following objectives;

• Describe the distribution of the variables that evaluate the performance of thesmall health food shop on the Sunshine
• Determine whether the sales are equal in all the seasons.
• Determine whether the profit differs from season to season.
• Determine whether the profit differs in different months.
• Give recommendation based on the findings.

The analysis focused on testing the hypotheses:

1. H₀: the average performance of different methods of payment is equal.

H₁: the average performance of different methods of payment is equal.

1. H₀: there is no difference in the average sale of different product location

H₁: there is a significant sales average difference by the product location.

1. H₀: the average net profit accrued from different selling position is not significantly different

H₁: the average net profit accrued from different selling position is significantly different.

1. H₀: The net sales in different seasons of the year were not different.

H₁: The net sales in different seasons of the year were different.

The analysis was designed to give a detailed discussion of the findings, obtained when assessing the given hypothesis.

Descriptive statistics for the net sales, average sales, profit total and cash total, were computed. The results are as summarized in Table 1.

Table 1: Descriptive statistics

Descriptive Statistics
	N	Range	Minimum	Maximum	Mean	Std. Deviation	Skewness
Statistic	Statistic	Statistic	Statistic	Statistic	Statistic	Statistic	Std. Error
Net_Sales	366	2370	0	2370	1014.26	313.986	-.362	.128
Average_Sale	358	53	8	61	18.52	3.985	3.758	.129
Profit Total	366	305.95	-33.98	271.97	30.7098	30.05661	3.274	.128
Cash_Total	366	1195	0	1195	404.29	153.643	.490	.128
Valid N (listwise)	358

The average net sale is $1,014.26 (SD = $313.99). The minimum net sale zero and the maximum of $2,370, meaning that the range is $2,370. The net sale has a negatively skewed plot, but it is not that much skewed. The average sales mean is $18.52 (SD = $3.99). The skewness coefficient shows that the average sales data are very skewed since the coefficient is greater than 2.00 (Kim, 2013). The same distribution is seen in the profit total, which has an average of $30.06 (SD = $3.27). Lastly, the cash total average is $404.29 (SD = $153.64). The data distribution is not that skewed since it is close to zero (Blanca, et al., 2013).

Research hypothesis

Figure 1 shows that the data are relative normally distributed around the measure of central tendency. However, there are a few outliers on both side of the tails.

The net sales box plot shows that the data are relatively spread around the median. This is indicated by a plot that has box and whicker that are almost equal on both sides.

The average sales box plot has a lot of outliers on the upper side. This supports the claim that the data are positively skewed.

The profit total box plot indicate that the data are positively skewed as there is a long tail to the higher side of the plot. Also, the outliers exist on the higher side of the box plot.

The first research question was assessed, where the means of the different payment methods were assessed. At a glance, the researchers evaluated the averages of each method, and the results are as follows.

Table 2: Mean report

Report
	Cash_Total	Credit_Total	Visa_Total	Mastercard_Total
Mean	404.29	584.80	555.85	22.09
N	366	366	366	366
Std. Deviation	153.643	228.860	244.870	67.823

The summary indicates that the total credit average is the highest (M = $584.80, SD = $228.860), followed by the Visa total average (M = $555.85, SD = 244.870) then the cash total average and lastly Mastercard total average is the least (M = $22.09, SD = 67.823). However, based on these statistics we cannot conclude or make inferences which method is doing the best or worst. Therefore, hypothesis test was carried to determine whether their averages were statistically different. Paired t-test was carried out for each of the pair methods of payment.

Table 3: Paired t-test (method of payment)

Paired Samples Test
	Paired Differences	t	df	Sig. (2-tailed)
Mean	Std. Deviation	Std. Error Mean	95% Confidence Interval of the Difference
Lower	Upper
Pair 1	Cash_Total – Credit_Total	-180.512	236.235	12.348	-204.794	-156.229	-14.618	365	.000
Pair 2	Cash_Total – Visa_Total	-151.563	249.173	13.024	-177.175	-125.950	-11.637	365	.000
Pair 3	Cash_Total – Mastercard_Total	382.192	170.933	8.935	364.622	399.763	42.776	365	.000
Pair 4	Credit_Total – Visa_Total	28.949	89.437	4.675	19.756	38.142	6.192	365	.000
Pair 5	Credit_Total – Mastercard_Total	562.704	236.975	12.387	538.346	587.063	45.427	365	.000
Pair 6	Visa_Total – Mastercard_Total	533.755	274.257	14.336	505.564	561.946	37.233	365	.000

The findings point that all the payment method averages were statistically different at the 95% level of significance. This suggests that the best performing method of payment is the credit, and the least performing method is the Mastercard. The results indicate that there is a difference in the amount yield by each of the four methods of payment. Thus, the business should emphasize the use of the most common mode of payment which is credit followed by the visa. The Mastercard seems to be doing very poorly, which can mean that it is not well recognized in this area.

Second, the research question was answered through testing the hypothesis: H₀: there is no difference in the average sale of different product location, Vs. H₁: there is a significant sales average difference by the product location. One-way ANOVA was carried out, and the findings are summarized in

Results and discussion

Table 4: One-way ANOVA

ANOVA
Total Sales ($)
	Sum of Squares	df	Mean Square	F	Sig.
Between Groups	134299725.024	4	33574931.256	37.176	.000
Within Groups	929333380.817	1029	903142.255
Total	1063633105.841	1033

The p-value shows that there is enough evidence to reject the null hypothesis (F (4,1029) = 37.176, p-value < .05) (Afifi & Azen, 2014). This means that at least one of the average sale is statistically different than that of the different places. The enterprise should not that different product position perform differently than the others. The post-hoc analysis indicates that rare and front are not performing differently, and also right and left (see Appendix 1). At the 95% confidence level, we can state that the sales of the Front position, Outside front position, and Right are significantly different. In a matter of fact, the outside front average is significantly higher ($3,384.37). Based, on these results, it is important to test whether the profit yield from the sale of the different positions differs significantly.

The hypothesis tested was whether the average net profit accrued from different selling position is significantly different. The summary of the one-way ANOVA is as follows.

Table 5: ANOVA table (net profit by sales position)

ANOVA
Net Profit ($)
	Sum of Squares	df	Mean Square	F	Sig.
Between Groups	36561758.739	4	9140439.685	46.211	.000
Within Groups	203534122.514	1029	197797.981
Total	240095881.253	1033

There is sufficient evidence to infer that the averages are significantly different (F(4,1029) = 46.211, p-value < .05 (Keller, 2014). The CEO of Harvest Kitchen should note that the outside front position of the enterprise makes the highest net profit. Similar finding like those of the average sales was found on the net profit as per the location of the business.

The net sales in different seasons of the year were assessed on whether they differ. The one-way ANOVA was carried out, and the results are as follows.

Table 6: One-way ANOVA (average net sales by season)

ANOVA
Net_Sales
	Sum of Squares	Df	Mean Square	F	Sig.
Between Groups	487761.038	3	162587.013	1.658	.176
Within Groups	35496528.544	362	98056.709
Total	35984289.582	365

The summary indicates that the research will fail to reject the null hypothesis (F(3, 362= 1.658, p =0.176). This means that the average sales for different seasons are not significantly different at the level .05 (Keller, 2014). This can be illustrated by the error bar below.

The 95% confidence interval on the error bar chart overlaps. This supports the claim that the averages are not statistically different. Therefore, we are 95% confident that the average net sales in the four seasons are not statistically different.

Further, an assessment was carried out to determine whether there is a significant difference in the average profit in the four seasons. The one-way ANOVA was carried out, and the results are as tabulated in

Table 7: ANOVA table (average Profit by season)

ANOVA
Profit Total
	Sum of Squares	df	Mean Square	F	Sig.
Between Groups	28591.757	3	9530.586	11.456	.000
Within Groups	301149.197	362	831.904
Total	329740.954	365

It is evident that the null hypothesis should be rejected (F(3, 362) = 11.456, p-value < .05) (Gelman, et al., 2014). The error bar was portrayed to identify which season had a significant total profit.

The error bars do not overlap, which indicates that the average of the profit totals in different seasons is statistically different (Hoekstra, et al., 2014). Therefore, the management of the small health food shop on the Sunshine Coast should know that different seasons yield a different profit. The most profitable season is the spring and the least profitable season is Autumn. This means that although the sales in the four seasons are not statistically different, the profit differs significantly.

Sales performance in different months

It was deemed important to test whether the average net sale is significantly different in different months of the year. The ANOVA test result is as follows.

ANOVA Table
	Sum of Squares	df	Mean Square	F	Sig.
Net_Sales * Month of the year	Between Groups	(Combined)	1399993.227	11	127272.112	1.303	.221
Within Groups	34584296.355	354	97695.752
Total	35984289.582	365

The average net sale is not significantly different in all the months of the year (F (11,354) = 1.303, p-value = .221) (Afifi & Azen, 2014). This means that it is expected the net sales is expected not to deviate so much, or to be different in all the months at the level .05.

Relationship between rainfall and profit

It is important to determine whether rainfall significantly affects the total profit earned. A linear regression model was fitted, and the results are as follows.

Model Summaryb
Model	R	R Square	Adjusted R Square	Std. Error of the Estimate
1	.008a	.000	-.003	30.13165
a. Predictors: (Constant), Rainfall
b. Dependent Variable: Profit Total

ANOVAa
Model	Sum of Squares	df	Mean Square	F	Sig.
1	Regression	19.078	1	19.078	.021	.885b
Residual	329573.612	363	907.916
Total	329592.689	364
a. Dependent Variable: Profit Total
b. Predictors: (Constant), Rainfall

The developed model is not significant (F (1, 363) = 0.021, p = .885). This implies that the model is not ideal in predicting the total profit earned. Therefore, we are confident that the rainfall amount received does not affect the profit of the business. The model cannot take into account any source of variation of the total profit when the rainfall is used as the predictor.

The standardized residual plot shows a trend. This indicates that the model is not ideal or there is no association between the profit total and the rainfall (Faraway, 2016).

Association between Product Class and the location in the shop

It was determined that the average net sale and total profit differ from one location of the shop to the other. Therefore, we need to assess whether there is an association between the location of the shop and the product class. This will help in answering the question which product class, located in which part of the shop gives the highest return. The chi-square test was carried out and the results are as follows.

Chi-Square Tests
	Value	df	Asymptotic Significance (2-sided)
Pearson Chi-Square	2957.360a	136	.000
Likelihood Ratio	2572.370	136	.000
N of Valid Cases	1034
a. 115 cells (65.7%) have expected count less than 5. The minimum expected count is .01.

There is adequate evidence to reject the null hypothesis (²(136, N = 1034) = 2957.360, p <.05) (Keller, 2014). This shows that there is consistency in where the products are placed in the shop. Therefore, this might be the main reason of the difference in the profit and net sales made on different location of the shop.

Conclusion

The report pointed a number of crucial findings. First, the outside front position makes the largest profit, and has the highest sales compared to all the other positions. It was established that the profit made from the sale from these positions is significantly different in some positions. This implies that there are some positions in the shop that makes more than the others. For instance, the outside front position makes the highest profit and sales that all the other positions. Therefore, the management should consider investing more in this area or increasing the commodities, both slow and fast moving, commodities. The net sales in the four seasons were not significantly different. However, the total profit average in the four seasons showed a significant difference. This means that although the net sale is not different, the profit in the Spring is significantly higher and in Autumn is significantly low. It is within the mandate of the business to establish what might be the cause of the difference in the profit since the net sales are still the same. This might be an increase in either overhead cost, increased the cost of commodities (without changing the profit margin), among others.

The results indicated that the average total profit obtained in different months of the year were not significantly different. This shows that the profit averages in all the months do not significantly deviate. The developed linear model indicated that rainfall was not a good predictor of the total profit earned. In other words, the rainfall does not affect the amount of profit the shop earns. It was found that the product class, was associated with where the product is located in the shop. This suggests that the difference in the profit and net sales made in different locations of the shop could be due to product difference in the locations. In particular, some products placed in a given location are fast-moving, and give a higher return.

To increase the profit margin, or maintain higher profit total throughout the year few things need to be considered. First, it should increase the outside front view of the business enterprise since this position makes the highest profit/sales. The institution should also, increase the Mastercard usage, and also accept cash, since these are the least used method of payment. The business should investigate which factors reduce the profit during the autumn, since the profit during this season is very low, but the net sales are still the same throughout the years.

References

Afifi, A. A. & Azen, S. P., 2014. Statistical analysis: a computer oriented approach. s.l.:Academic press.

Blanca, M. J. et al., 2013. Skewness and kurtosis in real data samples. Methodology.

Faraway, J. J., 2016. Extending the linear model with R: generalized linear, mixed effects and nonparametric regression models. Volume 124.

Gelman, A. et al., 2014. Bayesian data analysis, Boca Raton: CRC press.

Hoekstra, R., Morey, R. D., Rouder, J. N. & Wagenmakers, E.-J., 2014. Robust misinterpretation of confidence intervals.. Psychonomic bulletin & review, 21(5), pp. 1157-1164.

Keller, G., 2014. Statistics for management and economics. s.l.:Nelson Education.

Kim, H.-Y., 2013. Statistical notes for clinical researchers: assessing normal distribution (2) using skewness and kurtosis. Restorative Dentistry & Endodontics, 30(1), p. 52.

Leech, N. L., Barrett, K. C. & Morgan, G. A., 2014. IBM SPSS for intermediate statistics: Use and interpretation. s.l.:Routledge.