I have a structured interview to test a hypothesis, one of the questions required an answer from 1 to 5 .. I don't know how to analyze this? any recommendations?
@qulitative analysis
@Structured interview
There was an error in my original posting. It should have said, "In terms of doing a quantitative study, the main concern I would have is the size of your sample." With a small sample of experts, you be limited to doing descriptive statistics (e.g., percentages), but that may be good enough for your purposes.
Because there have been so many questions here about Likert-scored items and creating scales, I have created a set of resources on this topic:
https://www.researchgate.net/post/How_can_one_help_people_create_simple_scales_from_Likert-scored_items
Note that a Likert scale, even if numerical, has a clear qualitative flavour. In any case, could you provide more details?
Antonio M. Oller-Marcén thanks for your recommendation
Yes, I have two parts of my interview. Part one about construction project performance and the other part about the Maturity level of the construction company
I asked the experts to answer the first part from 1 to 5 for many factors which affect project performance. The other part I will measure it using a model. Then I will link them narratively.
and this is an example of the interview.
The only question in your interview that produces qualitative data is the final, open-ended item, and your example suggests that it could be answered in two or three sentences. Hence, there will not be enough data to conduct a full-scale qualitative study..
In terms of doing a quantitative study, the main concern I would have is the size of your sample. In particular, if you want to do hypothesis testing, you will need well over 100 participants in order to have enough "power" to produce meaningful tests of statistical significance.
David L Morgan , Maybe I used interview research method because the subject not well known in my area and I should do interviews with experts. The second part is about this subject and its a model to calculate the company's maturity level. Could and interview for experts be a tool for quantitative research?
There was an error in my original posting. It should have said, "In terms of doing a quantitative study, the main concern I would have is the size of your sample." With a small sample of experts, you be limited to doing descriptive statistics (e.g., percentages), but that may be good enough for your purposes.
Because there have been so many questions here about Likert-scored items and creating scales, I have created a set of resources on this topic:
https://www.researchgate.net/post/How_can_one_help_people_create_simple_scales_from_Likert-scored_items
Hello, Yasmin,
The short answer is yes, you could use a Likert-scoring or scale to code your answers, except for the last one that is open ended (for which you'll need to develop a content analysis scheme). Here is an important tip. When you're asking questions about, for example, age, income, years on a job, number of employees, always group your prospective responses in ranges. For example, 1.) 0-1 year, 2.) 1-5 years, 3.) 5-10 years, 4.) 10-15 years, 5.) more than 15 years. Thus when you're coding and analyzing your data you can say, "Thirty percent have one to five years job experience." If you organize your questionnaire that way respondents can check the range, saving you the difficult task of doing it yourself later. Make sense?
James Riggle Thanks for this tip. Yes of course it makes sense. I will do this in my data analysis chapter
But what do you recommend for analyzing part c? likert will be for each question for the whole sample, I want something for the whole part
Yasmin, for Part C as a whole I'd suggest you take the mean for all responses for each question, and the standard deviation to see how accurately the mean fits, summarizes the responses; I think you have a relatively small N, correct? Then you should figure out how best to fit Part C responses with Part A, for example. Are there different levels of "agreement" on Part C questions among respondents with 15 or more years of experience compared to those with 3 or less? Differences among managers of large projects, 40 or more employees, versus small ones, 10 or fewer employees. As a masters student you want to get comfortable, familiar with your data. Look for patterns in it (see the questions above). Soon enough your data will begin to "talk" to you, and your job in your thesis is to tell the story, what your data are saying, to your audience.
- Badges
- Science method
- Similar topics
- Methodology
- Research Methodology
- Qualitative Research