Scope of Research Sampling Method

44 2. Classical Test Assumption

a. Normality Test

According to Zulkifli Matondang 2009, normality tests are conducted in purpose to detect whether a set of data will be used as basic start to test hypothesis is empirical data that meets the naturalistic nature. Naturalistic nature is a thought that phenomena symptoms occur in this nature are natural and patterned. Widhiarso 2009 said that normality tests are some tests to measure whether our set of data having normality distribution so it can be used in parametric statistic. Tests of normality become important because this is a parametric test and have to normal distributed Haryadi and Winda, 2011. So, normality tests are some kind of tests to clarify whether the data obtained are normally distributed and, importantly, represent the whole population or not. Researcher choose two tools to test whether the data is distributed normally or not. 1 GraphAnalysis According to Ghozali 2006 normality test can use histogram graph by seeing the form of curve in the graph Normal Probability PlotP-P Plot namely with see at the spread of the data dots on the diagonal axis from the normal chart. Basic for decision-making are: 45 a For histogram graph, if the curve make a form of bell around the chart, so the regression model meet the normality assumption b For Normal Probability PlotP-P Plot, if the data spread around the diagonal line and follow the direction of the diagonal line, so the regression model meet the normality assumption. 2 Statistical Analysis Researcher uses tools of Lilliefors Kolmogorov- Smirnov because Haryadi and Winda 2011 suggested that if data of testing are more than 50 i.e. respondents are more than 50 people then use Lilliefors Kolmogorov- Smirnov test. Criteria for Lilliefors Kolmogorov- Smirnov test are: a Number of Kolmogorov- Smirnov significance Sig. 0.05, indicates the data normally distributed. b Number of Kolmogorov- Smirnov significance Sig. 0.05, indicates the data are not normally distributed NovitaItalianiKatsuri, 2011. b. Multicollinearity Test Multicollinearity test aims to test whether the regression model found a correlation between the independent variablesGhozali, 2009:95. A good regression model should not correlate between the independent variables. To detect the presence or absence of 46 multicollinearity in the regression model can be seen from the value of tolerance and the variance inflation factor VIF. Multicollinearity views of the tolerance value 0.10 or VIF 10. Both of these measurements indicate each independent variable which is explained by the other independent variables. c. Autocorrelation Test Autocorrelation is correlation between observed members arranged in time series if the data used is time series data or correlation among four contiguous variables Andriyatno, 2010. Diagnose the autocorrelation done through testing to test the value of Durbin Watson DW test by Ghozali2009:100.Here the criteria for testing autocorrelation. 1 If 0Dw DL there is any positive autocorrelation. 2 If DL Dw Du or 4-Du D 4-DL uncertain conclusion. 3 If 0 Dw DL or Du Dw 4-Du there is no autocorrelation. 4 If 4-DL Dw 4 there is any negative autocorrelation. d. Heteroscedasticity Test According to Ghozali 2009, the aim of heteroscedasticity test is to test whether the regression model occur the variance inequality of the residual from one observation to another observation. If the variance from residual of one observation to other observations is fixed, it is called homocedasticity andif it different called heteroscedasticity. A good regression model is homocesdasticity or