At the initial phase of a quantitative research, most of the researchers use qualitative field study to examine factors and variables and to affirm their relationships that are defined in the preliminary research model.
To gain a better understanding of the proposed relationship in a model, can we just start from quantitative study skipping the qualitative phase before?
If you have a good theory review about your topic, you can start with quantitative research. Start with qualitative research could help you to ajust your questionnaire for quantitaive data gethering.
Yes. Sequential Explanatory Design – This method is a two phase design where the quantitative data is collected first followed by qualitative data collection. The purpose is to use the qualitative results to further explain and interpret the findings from the quantitative phase. For example, a survey may be used to collect quantitative data from a larger group. Members of that group may then later be selected for interviews where they can explain and offer insights into their survey answers
Hi, In some cases, pilot study can be used with some experts or target respondents from the study population. Moreover, clear evidences from previous studies can be used to support your model.
Thanks Norbert for your suggestion. While having a qualitative study is a common way to adjust the questionnaire, but I am trying to find a strong justification on if someone has done the quantitative part first, is it still good?
Thanks Muhammad, What is someone does a concurrent study, collecting data from survey and interview later. Analysis for both collection done at the same time. The qualitative data trying to provide additional information for the quantitative data?
Thanks Ghazi...I still think if pilot can be sufficient to satisfy the reviewers?
There is no requirement to begin a quantitative study with a preliminary qualitative study. The common name for this approach is an "exploratory sequential" design, which points to the question of whether you need to do exploratory research in order to create your research instruments. So, the issue is whether you can satisfy your reviewers that existing measures are sufficient, and if not, then you face the issue of whether you can effectively create the measures yourself without any qualitative input.
As for sequential explanatory designs, the key issue there is using the qualitative data to understand the quantitative results. You can accomplish this with data that were collected concurrently, if you are careful to create one-ended items that can help you interpret the relationships between quantitative items. So, it is not such when the data are collected as it is how they are used.
Generally, if you have literature or other evidence (e.g. sales data, previous research) that the hypothesis and constructs under test are important and reasonable, reviewers won't indict you for skipping qual. After all, no one would ever consider qual in itself a conclusive validation of any grand model, so how could it ever be legitimate to use it to justify quant design? More often, it's about budgets and inertia.
A good mixed design, where you personalize the questionnaire through skips, branches, piping, randomized modules, etc. and include pointed open ends based on this personalization, will provide more value than a typical qual/quant [in either sequence] design. So I'd be more concerned about avoiding closed models so you can gain new insights, than qual/quant, per se.