Dear Ammar Orfahli ,

That sounds like poor planning. Didn't you write a proposal which included a methodology section? Don't you have research questions? You should consider how they can best be answered based around the questions you have asked and the data you have collected.

Another issue to consider is the size of your sample for your questionnaire. The larger it is, the more appropriate it is to emphasise quantitative methods.

Dear Ammar Orfahli ,

That sounds like poor planning. Didn't you write a proposal which included a methodology section? Don't you have research questions? You should consider how they can best be answered based around the questions you have asked and the data you have collected.

Another issue to consider is the size of your sample for your questionnaire. The larger it is, the more appropriate it is to emphasise quantitative methods.

Using questionnaire to collect data likely entails Likert scaled response and therefore ordinal data. Therefore you will use nonparametric statistical tools for the data analysis. That is, techniques like the sign test, Wilcoxson signed rank test, Kruskal-Wallis test, Spearman's rank correlation, etc.

Dear Peter Samuels ,

It was all a bit last minute because I have changed my topic at least 3 times in the past 4 months. I have not written a proposal. I do have research questions, and the interviews almost answer all of my questions, but I do not understand how to implement the data collected in my research. I have been told that I have to use some sort of model, and I do now know how or which to use. My supervisor hasn't been very useful so I lost hope in contacting him again.

You are using a mixed-methods research methodology. You may use SPSS to analyze the Quantitative part of your research. You could pick a suitable method to analyze the Qualitative part (using ATLAS or NVivo)

Two common mixed methods designs that you might consider are exploratory sequential (qual --> QUAN) or explanatory sequential (QUAN --> qual). In the exploratory design, you use your qualitative data to develop research questions (hypotheses) that you can examine with your survey data. In the explanatory design, you reach conclusions from you qualitative analyses that you can pursue in more depth and detail with your qualitative data.

Dear brother,

Humbly note that you ought to have had a proposal which when approved by your department must have research methodology section. You can request your department to change for you your supervisor if you cannot work together because she/he should be your mentor. At times students don’t meet time lines as they agreed with their mentors which can demoralize her/him. My suggestion is that you use descriptive survey research design in making a detailed examination. If the population under the study is homogeneous in nature where by both qualitative and quantitative research techniques is applied. Descriptive research involves gathering data that describe events and then organizes, tabulates, depicts, opinions, attitudes, or previous experiences through asking questions. Descriptive research can be used to obtain information concerning the current status of the phenomenon to describe what exists; with respect to specific variables or conditions in the situation. Describes descriptive research design as a systematic empirical inquiry into which the researcher does not have direct control of independent variables as their manifestation has already occurred or because they inherently cannot be manipulated. Descriptive design is appropriate since it enables the researcher to describe different scenarios, applied strategies.

According to Cohen et al (2007), descriptive research aims to contribute both to the practical concerns of people in an immediate problematic situation and to further the goals of research. The researcher therefore can seek to find answers to questions through analysis of various relationships of the factors that seemed to be associated with certain occurrences and an analysis of past events or already existing conditions.

You should have decided on your analysis in advance of designing your survey as your question design would be shaped by the analysis methods you want to use, i.e. if you want to use multivariate analysis you need scaled questions etc. If this has not happened you need to look at the type of questions that you have and then identify which analysis methods you could use. It's backwards and you may not get your research questions answered the way you need to. Stick to simple tests like chi squares if your data cannot be used for parametric analysis, it's better to do some simpler analysis correctly than use more advanced tools but with data that you should not use them for.

Are your interviews qualitative or quantitative. If they are qualitative, the most common approach for the kind of data is Thematic Analysis, Braun and Clarke (2006).

The statistical technique can be applied to analyze questionnaire survey data. However, selecting the right statistical technique should be based on your research questions. You can use SPSS software for statistical analysis. However, SPSS requires subscription fees. Alternatively, you can use JASP, which is freely available. There are some youtube tutorials. I like the tutorials of Dr Todd Daniel, which you can find using the link: https://www.youtube.com/watch?v=bVR3LrmfNYA&list=PLVI_iGT5ZuRnIqpav1UaadEJmj2e8QHzn.

You also have open-ended questions in your questionnaire as well as qualitative interviews with key personnel. These qualitative interviews can be analyzed through a thematic approach. However, the analysis of qualitative interviews should be based on your research questions. Do not waste your time if you don't have a particular research question.

Given that you've utilised both open-ended and closed questions, it'll be good for you to use software like SPSS for the closed questions and NVIVO for the open-ended / qualitative type questions.

With that sorted, you'll then be in a better position to apply specific analytical methodologies like Spearman ranking and simple correlation analysis for your quantitative or closed questions. For the more open-ended / closed questions, youll be able to generate themes from the NVIVO analytics, which can then be categorised or ranked, tailored to your research questions and objectives.

Ammar:

1. First, ask yourself what you want to study about the effects of Covid;

2. What are the relevant variables that you want to evaluate?

3. What are the attributes of the variables, and how do you want to measure (operationalize) the variables?

4. What economic theories do you want to investigate in order to accomplish your objectives? The theories should determine your models;

5. Estimate the models of preference for specific issues;

6. You have a range of options: Correlations as proposed by Emerson, Logit or Probit, structural VARs, Bayesian VARs (which I am currently exploring), and multivariate regression models with dummies (before and after Covid). Will you have enough observations (data points) after the outbreak of the pandemic? If not, that should shape your econometric decision.

NB: The World Bank, IMF, Bank of England, and FRED have very good and reliable time series data. I am usually suspicious of open-ended questions and their econometric value. What are the sample sizes of your surveys and what are the safeguards to guarantee the integrity of your data? Good luck!!

Ammar: Also see the data base of Johns Hopkins University for comparative data on the Covid pandemic.

Because you made a questionnaire with a mix of open ended and closed questions/answers conducted with interviews, you can use the Analytic hierarchy process technology.