RESULTS AND DISCUSSION Comparison of Parametric and Nonparametric Test Results for Two Correlated Samples
Two correlated samples pertaining to adult literacy-to-population ratio and employment-topopulation ratio were used to compare the test results derived from the parametric and nonparametric statistics. The succeeding discussion focuses not on the analysis of the SouthEast Asian demographics but rather on the comparison of the statistical outputs provided by the two said treatments.
The adult literacy rate as defined measures literacy among persons aged 15 years or older while the employment-to-population ratio refers to working-age population that is employed. For most countries, the working-age population is defined as persons aged 15 years or above (Table 1).
Table 1. South-East Asian Adult Literacy and Employment Ratios (Heyzer, 2013)
South-East Asian Countries Adult literacy-to- population ratio
(aged 15 and above) Employment-to- population ratio
(aged 15 and above)
Brunei Darussalam 95.4 63.5
Cambodia 73.9 81.4
Indonesia 92.8 63.2
Lao PDR 72.7 76.8
Malaysia 93.1 58.6
Myanmar 92.7 76.1
Philippines 95.4 60.1
Singapore 95.9 64.1
Thailand 93.5 71.1
Timor-Leste 58.3 54.5
Viet Nam 93.4 75.7
Since assessment of the normality assumption should be taken first into account before using the parametric test, the normality was examined visually using the normal probability plot. This method is a graphical way of assessing whether a set of data looks like it might come from a standard bell shaped curve, also known as normal distribution.
It must be noted that in a normal probability plot, the sorted data are plotted against values selected to make the resulting image look close to a straight line if the data are said to be approximately normally distributed. Deviations from a straight line indicate departures from normality.
The normal probability plot shown in Figure 1 suggests that most of the observations fall not so far from the line. This means that there is no strong indication of non-normality.
Figure 1. Normal Probability Plot for Adult Literacy and Employment Ratios
To further examine the nature of data distribution, the Kolmogorov-Smirnov test was computed to determine the degree of normality in the data. The value of this test provides a relative indication of normality, that is, as the value moves further away from zero, the data may be interpreted as it does not approximate a normal distribution. This test compares the set of scores in the sample to a normally distributed set of scores with the same mean and standard deviation.
Outcomes derived from Kolmogorov-Smirnov test posted a computed value of 0.1725, when compared to its critical value of 0.2844 at 0.05 level of significance, the data are interpreted as normally distributed (Table 2).
Table 2. Kolmogorov-Smirnov Normality Test for Adult Literacy and Employment Ratios
Demographic Kolmogorov-Smirnov Kolmogorov-Smirnov Conclusion
Variable Computed Value Critical Value at 0.05
Adult Literacy and Employment 0.1725 0.2844 Normally distributed
It is reasonable now to choose parametric statistics. To compare the two correlated samples, the parametric t-test was applied together with its nonparametric equivalent of Wilcoxon Τ test.
These treatments require two samples and it is necessary that those samples are to be paired. The said parametric and nonparametric tests are used when each observation in one group is paired with a related observation in the other group. Paired samples imply that each individual observation of one sample has a unique corresponding member in the other group.
It may be essential to note that the Wilcoxon T test has an advantage of using for it neither depends on the form of the data distribution nor on its parameters. It does not require any