collecting empirical data

G

hi guys. my masters dissertation has a lot to do with the effectiveness of business management systems. i would like to test the practicability of my hypothesis. how should i go about it? send questionairres to companies, go there personally, how many firms should i go to? should i ask my friends in business to fill them in & pass to their friends?

G

Are you expecting this to be publishable? If so, you can not pass forms to friends- There's a huge bias. If you go to firms, then you are collecting data using a form or interviewing. If interviewing you can get qualitative data. If youuse a check box you can get quantative data that you can code and analyze appropriately. Be sure the questions are asked in the same way each time, otherwise interviewer bias... Is your supervisor not helping you to design this?

G

no, we are supposed to do it on our own :(. How about i email the questionairres?

G

I take it you have not had a class in research design-that teaches how to make questionairres reliable, valid and eliminate as much bias as possible. It's important to have appropriate tools for testing the hypothesis. It isn't about how you deliver the questionairre- its about how the questionairre is designed. If you base a lot of your paper on a questionairre that is poorly designed (questions are worded in a manner considered Leading) it will get torn apart. Your supervisor should help you - not with the distribution of the materials, but with the design ofthe questionairre.

G

to hit the nail on the head, the issue here is not how to design a questionnaire or indeed how to come up with questions to ask my interviewees, the issue is : what the most effective method of getting adequate data is? interviewing? emailing questionnares? hand delivery of the questionnaires. no one needs a class in questionnaire design to have an opinion on this issue.

G

ps: when i mentioned handing them over to my friends, i meant people likely to have an opinion on the main issues covered therein that i know. my research is also looking at the effects of management systems on the community, so i suppose my friends are part of the community.

G

how about creating a website guys? is it worth it? am sure am not the first one on this site to do a project, so i'd appreciate some wisdom on those methods you've tried and tested. i'll be waiting.

G

George has some good advice. There are plenty of books on qualitative research methods and questionnaire design (and how to approach data collection) which will help with this sort of work. This isn't my field (give me nice quantitative science any day!) but I know that the big problem with questionnaires is getting people to fill them out and return them in the first place i.e. getting an adequate response rate. If you are going ahead with this type of data collection be prepared for a lot of time chasing people and making follow up phone calls ;o)

G

i already have the questionnares, so again, research design is not the issue. its what's the most effective way to use them. sorry if i sound like i don't appreciate your advice, i just don't feel like my question is being answered. i've done the door to door questionnaire before & it worked really well. am considering doing it again. the good thing is that they fill it there and then, so you don't have to chase them around. the terrible bit is gathering the nerves to go to them :).

G

after, giving it some thought, i have decided to do the door to door thing. i will get my questionnaire rechecked by some postgrads coz i have one of those supervisors who takes ages to give any response. i have based them on a design i used b4 for a research project during my undergrad. emailing is not gonna help i reckon, i'll let you know how i get on with the door to door thing. interviews..am looking for quantitative data so the questionnaires should be fine i suppose.

G

It may help to phone first and arrange a convenient time, people (especially businesses) often don't respond well to cold-calling.

G

ok, i will call them first and then see how it goes. thanks.

2222