An explanatory research is a research that tries to establish a connection between certain circumstances and a given phenomena. It usually studies a social behavior, occurrence or phenomenon and their potential causes; then concludes that the effects observed result of the causal factors identified. It is a cause-effect type of research, hence has a basis on scientific aspects of the used data. An example of an explanatory research is a study to determine whether increased abuse of drugs by youth leads to homelessness among them. In such a study, the researcher will try to link the occurrence of a phenomenon (youth homelessness) to a supposed cause (a rise in drug abuse among youth). Hedström (2004) has used explanatory research to explain the social stratification as a result of mobility or lack of it. The approach provides in depth information, hence developing comprehensive and sustained explanations to the findings.
Explanatory research designs have numerous benefits to researchers. From their nature, it can be seen that they can be used to test hypotheses. A researcher can use an explanatory research design to determine whether a claim is true or not. They can also be used in predicting the occurrence of future phenomena. For example, a researcher can use an explanatory research design to predict whether increasing cases of drug abuse among the youth will lead to a homelessness crisis. However, such a conclusion may require a goodness of fit test to become reliable. This future prediction of phenomena using explanatory research designs is possible due to its cause-effect nature which is not time-bound.
To test the relationship between the data collected and previous researches, a Chi-Square test or similar approach can be used to test for the existence of a statistically significant relationship (Miller & Siegmund, 1982). In this approach, the effects of changes in variables in the previous research are determined. Then, the effects of changes in similar or equivalent variables in the new data collected are studied as well. From these observations, a linear relationship may or may not be established. However, prior to decision making test for normality has to be factored to ensure that the analysis meets the statistical best fit.
Use of Chi-Square
A Chi-Square approach may not be appropriate for use in this research design. The Chi-Square method tests whether there is a statistical substantial relationship between the said variables; that is, youth homelessness and its supposed causal factors. If a Chi-Square test is used, sample data will be collected and checked for consistency with a hypothesized population. However, this research does not fit the criteria for suitability of a Chi-Square test: a simple random sampling might not work suitably, the variables involved are not categorical; and not more than five sample observations can be expected at each level of the variability. Although one or more of these conditions might be met, it is not likely that all of them are met and therefore that fact disqualifies the Chi-Square method as the most suitable approach in this research. If one or more goodness of fit unsatisfied conditions is ignored, the inferences arrived at are likely to be incorrect (McHugh, 2013).