the type of data analysis and even the software you use is highly dependent on your research questions and hypothesis. Are you going to get frequency or you are willing to measure the correlation between two variables?

I believe variables make more sense since I am looking for the factors that influence people's willingness to adopt electric vehicles, e.g.;

Environmental concern = more EV adoption willingness

Purchase cost = less EV adoption willingness

As far as I understood, you are facing two challenges. The first one regarding to the close ended questions and the second one arises from the open questions.

Firstly, if your are looking for the factors that influence people's willingness to adopt electric vehicles, a CB-SEM or PLS-SEM might be suitable aproaches to follow (apparently you're looking for a causal path and not for correlations between variables). Do you have obsevable variables to measure those factors (environmental concern, EV adoption willingness, purchase cost), right? If so, and depending on the sample size (desirably you should have more than 100 observations), advance with CB-SEM or PLS-SEM. If not, you have to based on descriptive statistics and eventually on correlations between observable variables to support your hypotheses (please be careful on their development because causal path differs from association/correlation between observable variables).

On the other hand, you should perform a content analysis for the open questions and coding the data (thematic analysis) taking into account the main goals of your study. You have several software to do this qualitative study: Maxqda, Nvivo or even Excel might be enough.

I hope it helps.

Good luck!

Hi Goncalo,

Many thanks for the answer. A few more questions;

1- Is it obligatory to specify dependant-independent variables in my thesis study?

2- Can you clarify what is CB-SEM or PLS-SEM (just the open versions, I will research them). Apologies, I have not used the method before.

3- I believe Surveymonkey will be enough as gives everything I need. Do I still need to code them etc.?

First, your statements indicate that you already have a dependent variable which you are trying to explain, EV adoption willingness. Also, you have some ideas about independent variables that might influence it, such as Environmental concern.

So, the next question is how you measured these concepts. For example, is EV adoption willingness represented by a series of questions, or just a single item?

Finally, CB-SEM or PLS-SEM are both advanced programs. I think you should concentrate on simpler approaches such as tables, t-Tests, and correlations.

David L Morgan Hi David, many thanks for the response.

EV adoption willingness - one direct question(e.g., how likely are you to move to an ev in future), two indirect ( e.g., do you think electric cars are the future?)

Last question for instance - I measure the variables, e.g. environmental concern ask I ask participants to choose whether they are very important, important, not important at all etc.

My question is actually whether I can go and analyse my questionnaire question by question? I can pull out a very detailed data from Survey Monkey and I do not think i will need an advanced program. What do you think?

If you want to investigate what affects EV adoption willingness, then you need to do more than examine your data one variable at a time. In particular, that research question ask how other variables are related to EV adoption willingness.

Depending on how many categories you have in each variable and your total sample size, it may be possible to examine the relationships among your variables with tables (cross tabulations).

You may use 'Statistical Package for the Social Sciences (SPSS)'

Dear Gökay Cinar if you are a masters student in the department of Pol Science, you can opt for an only qualitative study in which descriptive and qualitative analysis of your data would be good enough to support your discussions.

For a quantitative study, there is a huge difference between a general survey (e.g., EV adoption self created questions) and a constructs measurement survey.

You can make use of the Likert Scale, create your data file in an excel sheet, then use SPSS for the analysis and interpretations. Using SPSS, you can analyze the data with t-test, correlation, regression, ANOVA, or chi-square

Analyze it accordingly in line with your research methodology plan. You can easily analyze it by both qualitative and quantitative methods respectively.