55 and analysis of documents. So it can be seen that the time horizon used in
this research were researchers study time series, where this research is more emphasis on research of succession time data. The data collected can
also be directly downloaded through the website www.idx.co.id
, the researchers collected data information used as the basis for the theory and
books and literature-literature related to the preparation of this research.
D. Analysis Method and Testing of Hypothesis
Completion of this research by using quantitative analysis techniques. Quantitative analysis is used in a way to analyze a problem
that realized by quantitative. In this research, quantitative analysis is done by quantifying the research data to generate the information needed in the
analysis. The analytical tool used in this research is the logistic regression
analysis. The assumption of normal distribution can not be fulfilled because the independent variable is a mix between continuous variables
metric and categorical non-metric. In this case can be analyzed by logistic regression because it does not need the assumption of normality
test of data on independent variable. Stages in testing by using logistic regression test can be explained as follows Ghozali, 2013:
56
1. Descriptive Statistics
Descriptive statistics are used to provide a description of the data that is visible from the average, standard deviation, variance,
sum, range, kurtosis, maximum, minimum and skewness Ghozali, 2013; 19. Mean used to estimate the average size of the estimated
population of the sample. Standard deviation is used to assess the dispersion of the sample average. Minimum and Maximum used to
view the minimum and maximum values of the population. This needs to be done to see the overall picture of the samples were
collected and are eligible for the research sample.
2. Logistic Regression
Hypothesis testing is done by using a multivariate logistic regression. Logistic regression is used when the dependent variable
is a non-metric with two categories and the independent variables are one or more metric and non-metric Ghozali, 2013: 333. The
dependent variable in this research is a going concern audit opinion expressed by a dummy variable, where category 1 for the company
going concern and category 0 for companies non-going concern.
3. Overall Model Fit test
The first analysis done is to assess the overall fit model to the data. Hypotheses to assess model fit is:
H : hypothesized model fit to the data
H
a
: hypothesized model that does not fit with the data