Step 7- ANN (RQ2)
5. DATA ANALYSES AND RESULTS
5.9 Comparing Respondents on Demographic Factors
Demographic factors such as gender, age, education, and income were integrated within the study’s research model to identify respondents’ possible differences along these lines in influencing user continuance intention towards M-pesa. To determine possible differences within the studied population on demographic factors, the study’s data was submitted to an independent samples T-test 77and a one-way ANOVA78. A t-test examines whether a difference between two groups’ averages is improbable to have arisen due to random chance in sample selection. The t- test’s statistical significance and the t-test’s effect size are the core outputs of the t-test. Statistical significance shows whether the difference between sample averages is likely to represent an actual difference between populations, and the effect size indicates whether that difference is large enough to be practically important. Analyses of the differences in demographic factors on user continuance intention towards M-pesa are presented next.
77 A t-test is a statistical test that can determine if there is a significant difference between two groups on a dependent variable.
78 ANOVA is a statistical technique used to determine possible differences in group means within a population (Hair et al. 2014a, p.666).
5.9.1 Gender
The study’s data was submitted to an independent sample t-test to compare the average value of each group: males and females to the average value of the dependent variable (continuance intention). Thus, a requisite for a t-test is that the independent variable be nominal and consist of only two values and the dependent variable be a scale-level variable. The results of the t-test revealed that no significant difference exists between males and females in their responses towards continuance with M-pesa. The continuance variable consists four items (abbreviated as CONT) and responses by gender across these items are as follows:
CONT 1: t(432) = .79, p = .42; CONT 2: t(432) = .97, p=.32; CONT 3: t(432) = .59, p=.55;
CONT 4: t(432) = .35, p=.72. On a seven-point likert scale with one being strongly disagree to continue use of M-pesa and seven being strongly agree to continue use of M-pesa, males averaged on the four items: CONT 1: 6.36 (SD = .584); CONT 2: 6.31 (SD=.543); CONT 3: 6.07 (SD=6.87); CONT 4: 6.29 (SD=.572). Whereas, females averaged on the four items: CONT 1:
6.40 (SD=.509); CONT 2: 6.37(SD=.550); CONT3: 6.11(SD=.696); RCONT 4: 6.31(SD=.535).
A summary of the results of the t-test are presented in table 5.12 below.
Table 5.12 Independent Samples T-test
Group Statistics Independent Samples Test
Gender (Male-1 vs Female-2) Vs Continuance Intention
N Mean Standard Deviation Std.
Error Mean
F Sig. t df Sig.
(2-tailed)
Mean Differ ence
Std.
Error Differ ence CONT 1 1-M 204 6.36 .584 .041 1.820 .178 -.793 432 .428 -.042 .052
2- F 230 6.40 .509 .034
CONT 2 1- M 204 6.31 .543 .038 .640 .424 -.979 432 .328 -.051 .053
2- F 230 6.37 .550 .036
CONT 3 1- M 204 6.07 .687 .048 .474 .492 -.594 432 .553 -.040 .067
2- F 230 6.11 .696 .046
CONT 4 1- M 204 6.29 .572 .040 .282 .596 -.356 432 .722 -.019 .053
2- F 230 6.31 .535 .035
Note: A seven point scale was used to measure responses and ranged as follow: (i) Strongly disagree, (ii)Disagree, (iii)Somewhat Disagree, (iv) Neutral, (v) Somewhat Agree, (vi) Agree, (vii) Strongly Agree.
5.9.2 Age
The One-way ANOVA compares the means of the samples of more than two groups in order to make inferences about the population means. It tests the null hypothesis that a sample of more than two groups is drawn from an identical population. A one-way between subjects ANOVA was conducted to compare the effects of Age on user continuance intention towards M-pesa within age brackets: 18-30, 31-40, 41-50, and 50 and above. Continuance intention was reflected by four items, and the results for each are as follows: CONT 1: [F (3,430) = .666, p=.573], CONT 2: [F (3, 430) = .180, p=.910], CONT3: [F (3, 430) = .348, p=.790], CONT 4: [F (3, 430) = .321, p=.810]. Collectively, these results suggest that for the study’s sample, the age of individual users of M-pesa do not have an effect on continuance intention towards M-pesa, and a summary of the results are presented in table 5.13 below.
Table 5.13 One- Way Anova: Age and Continuance Age vs
Continuance Sum of
Squares df Mean
Square F Sig
CONT
1 Between
groups .595 3 .198 .666 .573
Within
Groups 128.144 430 .298
Total 128.740 433
CONT
2 Between
groups .162 3 .054 .180 .910
Within
Groups 129.368 430 .301
Total 129.530 433
CONT
3 Between
groups .502 3 .167 .348 .790
Within
Groups 206.625 430 .481
Total 207.127 433
CONT
4 Between
groups .295 3 .098 .321 .810
Within
Groups 131.558 430 .366
Total 131.853 433
5.9.3 Education
The second executed one-way between subjects ANOVA was conducted to compare the effects of Education on user continuance intention towards M-pesa within six education categories (primary, secondary, post-secondary school, university or college, postgraduate, and other).
Continuance intention was reflected by four items, the results for each are as follows: CONT 1: [F (5,428) = .887, p=.489], CONT 2: [F (5, 428) = .777, p=.567], CONT3: [F (5, 428) = .936, p=.457], CONT 4: [F (3, 428) = .474, p=.796]. Taken together, the above results suggest that for the study’s sample, the education levels of M-pesa users do not have an effect on continuance intention towards M-pesa. A summary of the results are presented in table 5.14 below.
Table 5.14 One-Way Anova- Education and Continuance Education vs
Continuance Sum of
Squares df Mean
Square F Sig
CONT
1 Between
groups 1.321 5 .264 .887 .489
Within
Groups 127.419 428 .298
Total 128.740 433
CONT
2 Between
groups 1.165 5 .233 .777 .567
Within
Groups 128.365 428 .300
Total 129.530 433
CONT
3 Between
groups 2.241 5 .448 .936 .457
Within
Groups 204.885 428 .479
Total 207.127 433
CONT
4 Between
groups .726 5 .145 .474 .796
Within
Groups 131.126 428 .306
Total 131.853 433
5.9.4 Income
A final one-way test between subjects ANOVA was conducted to compare the effects of income on user continuance intention towards M-pesa within five income brackets (less than 500,000, 501,000-1000, 000, 1001, 000-2,000,000, 2,001,000 – 3,000,000, and over 3,000,000).
Continuance intention was reflected by four items, the results for each are as follows: CONT 1: [F (4,421) = 1.610, p=.171], CONT 2: [F (4, 421) = .316, p=.867], CONT3: [F (4, 421) = .068,
p=.991], CONT 4: [F (4, 421) = .777, p=.541]. In sum, the above results suggest that for the study’s sample, income levels of M-pesa users do not have an effect on continuance intention towards M-pesa. A summary of the results are presented in table 5.15 below.
Table 5.15 One-Way Anova: Income and Continuance Income vs
Continuance Sum of
Squares df Mean
Square F Sig
CONT
1 Between
groups 1.911 4 .478 1.610 .171
Within
Groups 124.952 421 .297
Total 126.864 425
CONT
2 Between
groups .381 4 .095 .316 .867
Within
Groups 126.943 421 .302
Total 127.324 425
CONT
3 Between
groups .123 4 .031 .068 .991
Within
Groups 190.123 421 .452
Total 190.246 425
CONT
4 Between
groups .952 4 .238 .777 .541
Within
Groups 128.984 421 .306
Total 129.937 425
Given findings that no significant patterns exist within the data across respondents’ demographics, the researcher proceeds to determine the dimensionality, reliability, and validity of the constructs in the study.