Examining Users And Non-Users Of Self-Checkout Counters Essay.

Aim and Research Questions

In this part, you are asked to refer to and familiarize yourself with the research article entitled “An examination of users and non-users of self-checkout counters” and respond to the requirements below.

Explain briefly, the aim of this research article and state the research questions chosen by the authors of this research paper. 

The research aims to examine the consumers’ adoption of self-checkout technologies in Singapore. This is through comparing users and non-users of self-service technologies and situational factors encouraging their use. The objective of this is to provide an understanding of the differences between users and non-users of self-checkout counters and help retailers devise strategies to encourage the people who were non-users to adopt the use of self-checkout counters. This was also to help retailers leverage self-checkout technologies to raise the level of productivity and reduce the reliance on labour.

Referring to the “Survey instrument” used in this study, explain how this instrument was organised and name each section. 

The survey instrument consisted of three sections. The first section measured whether the respondent was a user of self-checkout counters by the rating the frequency of use of self-check-out counters when available at the stores. The second section had a 5-point Likert scale with questions measuring perceptions of self-checkout counters and their relative advantage over staffed cashiers, ease of use, reliability, and entertainment value. The compatibility of self-checkout counters with the respondents’ lifestyles was also measured. The third section had a 7-point Likert scale statement where respondents indicated the likelihood of using self-checkout counters in different situations.

b. Refer to the second section of this instrument, name the concept that is measured and in how many factors it was subdivided. How many items form each factor?

The section is measuring the levels of internal consistency of the survey instrument. This is done through a 3-item scale where the dimensions of Relative Advantage, Perceived Complexity, Reliability and Fun were measured using the Cronbach α coefficients

c. The article refers to the Cronbach coefficients. What do these represent and why it is important to include them in this paper?

Cronbach’s coefficient is a measure of internal consistency. This shows how items are closely related in a group. This is also referred to as scale reliability (Groebner, Shannon, and Smith, 2013). It assumes values of between 0 and 1. Multiple questions on a Likert scale in a survey a checked whether reliable and tell the closeness of items in a group. High values of the coefficient may mean a high correlation. However, the number of items should be checked since many may result in high alpha values. A low alpha value may suggest few questions on the test hence can be improved by adding more relevant items.

Name and explain the sampling method used for the collection of data. Explain whether the responses were random and independent by giving evidence from the text.

The convenience sampling method was used in data collection. This is a non-probabilistic method of sampling where data is collected from the convenience of a pool of available respondents. The sampling method is non-random. The convenience sample was taken from shoppers and residents within the residential suburb. The data collection period was from 21 January to 7 February 2014. Since it was under the convenience of the researcher, this should non-randomness.

Survey Instrument and Internal Consistency

a. Referring to the section “Demographic variables in the use of self-checkout counters”, the authors have concluded that there is no difference in the use of self-checkout counters across different demographic segments. What evidence have they given to evidence this conclusion?

According to the Chi-square tests for independence to examine any association between the demographic variables, a non-significant association between gender and the use of self-checkout counters, χ2 (1, n = 778) = 1.32, p = 0.25, ? = −0.04), age (χ2 (4, n = 778) = 4.06, p = 0.40, ? = 0.07) and education (χ2 (3,n = 778) = 4.26, p = 0.24, ? = 0.07) with the use of self-checkout counters. Since resultant p-values are >.05 the demographic segments were not significant at 5% level of significance.

b. In the same section, the report for the chi-squared test against the variable “Age” reads “²(4) = 4.06, = 0.40, = 0.07”. What is “4” in this report and explain how it was calculated? 

The 4 represents the degrees of freedom of the age variable. It is normally taken as df=n-1. Where n is the sample size. There are 5 age groups. Therefore, df=5-1=4.

Referring to the section “Evaluation of self-checkout counters”, an independent t-test was used to compare the average scores for all the measures between users and non-users. Explain what could be the reason that all the test results were found to be statistically significant but with moderate effect sizes.

The statistical significance alone could be misleading since it is influenced by the sample size. On the other hand, the effect size is independent of the sample size (Cohen,1988). The effect size, therefore, shows the practical significance of the test

5. Conclusion And Further Research

The authors referred to a “recent” study. What does this study suggest and what future research could be conducted to test it? 

In a recent study, the reason for not using the check-out counters is attributed to the reluctance or resistance to adopting the technology. The degrees of resistance is said to be different (Patsiotis, Hughes, and Webber 2013) Further research should be conducted to examine whether there are differences between non-users of self-checkout counters and to further isolate strategies specific to each segment of non-users.

For this case study, we are required to produce the following tasks

1. Produce a frequency table for the dependent variable “Take up” and report your findings.

According to the results above,55 (63.2%) agreed to consider shopping at AMAZON Fresh out of the total sample size of 87 subjects. However,32 (36.8%) did not consider shopping at AMAZON Fresh.

2. Conduct an appropriate hypothesis test fully to inform on the RQ1 above

Null hypothesis (Ho):  The proportion of people who consider shopping at Amazon Fresh is less or equal to 50%

Alternative hypothesis (Ha): The proportion of people who consider shopping at Amazon Fresh is greater than 50%.

The group that was considered shopping at Amazon Fresh had an observed proportion of (.63, N=55) while those that did not consider shopping at Amazon Fresh had a proportion of (.37, N=32) with a significant value of p=.018 according to the binomial test which is significant. We, therefore, reject the null hypothesis and conclude that the proportion of people who consider shopping at Amazon Fresh is greater than 50% (Johnson and Bhattacharyya,2019).

Sampling Method

3. Conduct an appropriate hypothesis test fully to inform on the RQ2 above

Null hypothesis (Ho): The proportion of people who consider shopping at Amazon Fresh is similar between those who are comfortable with mobile technology and those who are not. 

Alternative hypothesis (Ha): The proportion of people who consider shopping at Amazon Fresh is different between those who are comfortable with mobile technology and those who are not.

The Chi-Square Test of association was performed to assess the relationship between Amazon Fresh take up and mobile technology. There was a significant relationship between the two variables,(1, N=87) = 5.120, p = .024. We reject the null hypothesis and conclude that the proportion of people who consider shopping at Amazon Fresh is different between those who are comfortable with mobile technology and those who are not.

4. General question: Explain clearly and in your own words the reason for conducting a hypothesis test instead of simply comparing sample statistics of RQ2 above

Hypothesis testing is used for investigating the ideas that exist in the world of statistics by researchers. This process is important in forming and testing solutions to problems. It is based on the prediction of what research will find out. This allows conducting a scientific study that will either reveal that the hypothesis is true (by rejecting the null hypothesis) or false by accepting the null hypothesis. The use of sample statistics alone is not scientific and cannot be retested by a different researcher who can replicate the study to find out if is true or not.

1. Produce the descriptive statistics for the entrepreneurship intention scores split by the two groups of students, those who studied for an entrepreneurship module and those who didn’t and test the normality and the homogeneity of variance assumptions. Report your findings

The descriptive statistics reveal that the sample of 74 students who studied the entrepreneurship module had a slightly higher score of about 1.313 units. The group had mean scores and standard deviation of (M=22.23, S.D=3.041) and a median score of 22 units. The sample of 60 students who did not study the entrepreneurship module had a slightly mean score and standard deviation of (M=20.917, S.D=2.625) and a median score of 21 units. The Levene statistic of 1.039 and p=.310 implies the homogeneity variance assumption is satisfied. The Kolmogorov–Smirnov test (n ≥50) is a non-significant (p=.202) hence, the normality assumption is satisfied (King’oriah,2004).

2. Test if the average scores for entrepreneurship intention are significantly different and report your findings. [Max: 50 words]

Null hypothesis H0: There is no difference in mean scores of the two groups of students under the entrepreneurship intention program.

Alternative hypothesis Ha: There is a significant difference in mean scores of the two groups of students in the entrepreneurship intention program.

The difference was significant t(132)=-2.64,p=.009.Since p<.05, the null hypothesis is rejected. It is concluded that there is a significant difference in mean scores of the two groups of students in the entrepreneurship intention program (Siegel, 2016).

3. Test if the university department has an effect on the average score for entrepreneurship intention and report your findings. 

Null hypothesis H0: The university department has no effect on the average score for entrepreneurship intention

Alternative hypothesis Ha: The university department has an effect on the average score for entrepreneurship intention

The group mean and standard deviation score for GSBL was slightly higher (M=22.87, S.D=2.872) than CASS (M=21.409,S.D=2.653) and CDM (M=20.591, S.D=2.831). The Anova resulted in a F(2,131)=7.738,p<.001which indicates a significant result. We reject the null hypothesis and conclude that university department has a significant effect on the average score for entrepreneurship intention (Newbold,Carlson and Thorne, 2013).

4. General question: Explain in your words the concept of the p-value in hypothesis testing, how it is used and what is its limitation.  

P-value shows the level of marginal significance and the probability of an event occurring. It provides a value that the researcher compares with the level of significance. When the p-value is less than the chosen level of significance, the null hypothesis is rejected while the null hypothesis is accepted when the p-value is greater than the level of significance (Berenson, et.al.,2012). However, since this value depends on the sample size, a significant result can be got when a large sample is involved and a non-significant result when the sample size is small regardless of whether the effect size of the data is large.

References 

Berenson, M., Levine, D., Szabat, K.A. and Krehbiel, T.C., 2012. Basic business statistics: Concepts and applications. Pearson higher education AU.

Cohen, J., 1988. The concepts of power analysis. Statistical power analysis for the behavioral sciences, 1, p.17.

Groebner, D.F., Shannon, P.W., Fry, P.C. and Smith, K.D., 2013. Business statistics. Pearson Education UK.

Johnson, R.A. and Bhattacharyya, G.K., 2019. Statistics: principles and methods. John Wiley & Sons.

King’oriah, G.K., 2004. Fundamentals of applied statistics. Jomo Kenyatta Foundation, Nairobi, Kenya.

Newbold, P., Carlson, W.L. and Thorne, B., 2013. Statistics for business and economics. Boston, MA: Pearson.

Patsiotis, A.G., Hughes, T. and Webber, D.J., 2013. An examination of consumers' resistance to computer-based technologies. Journal of Services Marketing.

Siegel, A., 2016. Practical business statistics. Academic Press.