Really concerned about my thesis

G86

Hey Guys,

I'd really appreciate any advice here.

I'm working on a thesis which is due in at the end of the month, and I've yet to do my methodology. It's qualitative and I'm doing semi structured interviews but I'm pretty clueless on how I go about analysing the data and to be honest just don't know where to start. I searched the net etc. and came up with perhaps using content analysis but I'm not sure. I did really well in the Lit Review part of it which has been graded by my tutor, but I've only had one meeting with her all year and we've been told that the tutors are uncontactable over the summer so I'm not sure what to do. We did have a sub for a week during the year who was great so I was hoping to maybe go to her from grinds or something in qual methods but she's out of the country.

I worked so hard this year and put so much time and effort into it as well as trying to keep my head above water financially. I was sitting comfortably enough on a 2:1, with maybe a chance of a first, but now I'm really concerned that the thesis is going to pull me way down as it contributes to over a third of my result; I just don't want the past year to have been for nothing I'm not sure what to do to be honest.

I've emailed my tutor on the off chance she'll check her mails, and also another lecturer, so hoping that someone gets back to me. But if not then I'm kind of at a loss to be honest.

Thanks for any suggestions,

Grá

LegacyUser

Hiya, you sound pretty stressed out, I haven't done a thesis myself yet so I don't know how incredibly worried you must be feeling. You seem to have been working well during the year so going by that i'd say your doing much better than you think you are. Maybe leave a message on boards forum of your college you may find more people can help you there

G86

Thanks for the reply, but I just got an email from my tutor to say she can't help during the summer months I've been talking to a few of my classmates but they don't seem to have a clue either, they don't seem overly stressed - but I still can't help being worried that I'm going to do it all wrong..

Fremen

If your department has any PhD students, you could send an e-mail to the department secretary asking for extra tuition (say twenty quid an hour), and she could forward it on to them.

If it's quantitative stuff you're having trouble with, you could try the maths/stats dept too.

If there are no PhD students at your uni, try somewhere nearby that does have them. We get random emails asking for stuff like this all the time at my college.

Sinall

I have just finished mine and have some good info/reading material/copies of other theses. PM me if you want these!

I did mine part time in DCU (as I am working full time) and the methodology was hard. You have loads of time though. I did quantitative research, supported by qualitative semi-structured interviews, so a lot of the stuff I had would be very relevant to you.

Miaireland

I had a similar problem when I was doing my thesis and ended up googling. Lots of the US university have course literature online that give you step by step details of what to do.

Big inner

Hi Op,

In similar situation myself in terms of time frame but during my meetings with my Tutor, he suggested i foucs on themes that come from my interviews. i.e. identify common themes that are linked to your objectives.

I have yet got to this stage or the interviews (however they are organised, phew) so not too sure how they will pan. Anyway best of luck, roll on Sept!

G86

Hi Guys

Really appreciate the replies It's qualitative research, it's a DCU degree but I completed it in a smaller college and the whole college is closed at the moment, I don't really have many contacts in DCU itself. I was thinking about running with themes, but after googling it I got a bit worried as all of the info seems to advise using a certain form of analysis, e.g. content.

I know it seems likea fair amount of time, it's just I work 5 days a week so I know myself I'm a bit limited time wise now.

Sinall

Mine was due in today!!! My own supervisor was on annual leave for the last while, but if you pop out to DCU (if you are in Dublin) there are usually other lecturers around who can answer questions.

I can give you some DCU contacts if you like.

Ostrom

PM sent Gra - content analysis is not suited to interview data. There are some excellent texts on interview analysis that deal with grounded theory and coding, which would work much better.

These books are available in DCU library, and would be a good starting point:

Kvale, Steinar. 2007. ‘Doing Interviews’. London: Sage

Kvale, Steinar & Brinkman, Svend. 2009. Interviews: Learning the Craft of Qualitative Interviewing. London: Sage.

Schutt, Russell K. 2006. Investigating the Social World: the Process and Practice of Research. London: Sage (Chapter 9, Qualitative Methods)

Lollipops23

OP i just PMd you.might be of interest;)

PK2008

Ok I dont know how DCU work it but you should have had to submit your research methodology before you started your research otherwise how can you be sure you are approaching it in the right way??

It sounds like you've done some data collection and are now trying to fit a methodology around it, if so, then i assume you havent formulated a hypothesis yet.

For qualititate research where a hypothesis hasnt been formulated (exploratory) then "Grounded theory" is a apopular methodology.

Look it up but basically it is is a methodology which starts out by collecting qualitaitve data then coding it, then formulating a hypothesis based on the recurring "themes" in the data.

Lollipops23

PK2008 wrote: »

Ok I dont know how DCU work it but you should have had to submit your research methodology before you started your research otherwise how can you be sure you are approaching it in the right way??

It sounds like you've done some data collection and are now trying to fit a methodology around it, if so, then i assume you havent formulated a hypothesis yet.

For qualititate research where a hypothesis hasnt been formulated (exploratory) then "Grounded theory" is a apopular methodology.

Look it up but basically it is is a methodology which starts out by collecting qualitaitve data then coding it, then formulating a hypothesis based on the recurring "themes" in the data.

I'm in the OPs year actually, we had to write up a research proposal including our intended methodology. for instance,my lit review was slated but my meth. was fine.

PK2008

Lollipops23 wrote: »

I'm in the OPs year actually, we had to write up a research proposal including our intended methodology. for instance,my lit review was slated but my meth. was fine.

Well then the OP should use the methodology they submitted in the proposal

(and you should choose some different literature to review)

Sinall

PK2008 wrote: »

Well then the OP should use the methodology they submitted in the proposal

(and you should choose some different literature to review)

Not necessarily. If you are happy with your lit review and you have discussed with your supervisor the most appropriate research strategy, or if from your lit review, one type of research is most relevant, then you can go ahead and do your research and actually formulate your methodology chapter afterwards.

Depending on your research topic/question, you will know whether one type of research is more suitable than another. Some research questions will be more suited to qual, some to quant and some to a mixture.

PK2008

Sinall wrote: »

Not necessarily. If you are happy with your lit review and you have discussed with your supervisor the most appropriate research strategy, or if from your lit review, one type of research is most relevant, then you can go ahead and do your research and actually formulate your methodology chapter afterwards.

Depending on your research topic/question, you will know whether one type of research is more suitable than another. Some research questions will be more suited to qual, some to quant and some to a mixture.

Im not being smart but to formulate your research methodology after you do your research doesnt make any sence. Pretty much the whole point of putting together your research methodology is so you can show your supervisor that you are approaching your study in the appropriate manner. Your research methodology should be one of the first things you do- not the last.

At some point the OP must have decided that qualitative data and interviewing was the correct method to collect data to answer her research question- how did she come to that conclusion/ decision?? (that should have been part of her methodolgy justification)

Unless your doing exploratory research where a hypothesis cannot be formulated before you undertake some kind of data collection, but even in that case you would need to submit it as the most appropriate methodology based on an argument as to why the study is exploratory.

Like I said, I dont know how they do it in DCU but I just did a masters thesis for UCD and I had to submit my methodology last year (basically 6 months before the thesis was due), which had to be approved before I could even start the research. It didnt happen to me but anyone who selected the wrong methodology had to go back and start again.

OP - Id say you have to refer to th emethodolgoy you stated in your proposal- if you deviate from that then it could put your research into question. You really do need to speak to your supervisor

Sinall

PK2008 wrote: »

Im not being smart but to formulate your research methodology after you do your research doesnt make any sence. Pretty much the whole point of putting together your research methodology is so you can show your supervisor that you are approaching your study in the appropriate manner. Your research methodology should be one of the first things you do- not the last.

At some point the OP must have decided that qualitative data and interviewing was the correct method to collect data to answer her research question- how did she come to that conclusion/ decision?? (that should have been part of her methodolgy justification)

Unless your doing exploratory research where a hypothesis cannot be formulated before you undertake some kind of data collection, but even in that case you would need to submit it as the most appropriate methodology based on an argument as to why the study is exploratory.


Like I said, I dont know how they do it in DCU but I just did a masters thesis for UCD and I had to submit my methodology last year (basically 6 months before the thesis was due), which had to be approved before I could even start the research. It didnt happen to me but anyone who selected the wrong methodology had to go back and start again.

OP - Id say you have to refer to th emethodolgoy you stated in your proposal- if you deviate from that then it could put your research into question. You really do need to speak to your supervisor

And I'm not being smart either, but sometimes, from your lit review and understanding of what methods previous researchers have used, and from the type of research question that you have, you know that one type of research is going to be more appropriate than others. And there is always the chance that her supervisor indicated which research method would be most appropriate when she was discussing her lit review. Anyway, the OP seems set on a particular method of research and given that her results have been so good to date, I imagine that she has not just picked a research method by chance, but has given some thought and consideration to this. There is no point in telling the OP that she should have already done her methodology, the fact is it isn't done and she is doing it now and looking for help and advice. What works for one person might not work for another person. People take different approaches. And, yes, there is a preferred academic approach but sometimes, when you are working full time as well as studying (as is the OP and as was I) you don't always get to work to a preferred academic schedule. My thesis supervisor took this into account, but she was used to supervising part-time Masters students.

The OP has said that her supervisor has emailed back and said that she can't help her over the summer, which I don't think is fair, so I have PM'd her and sent her a DCU contact and am going to email her sample methodologies. I have today submitted my masters thesis in DCU so hope that my experience will be of practical benefit to the OP. If her supervisor is not willing to give her guidance at this stage then she has to rely on her own instincts and others assistance.

PK2008

Im not geting in an argument with you but if you read my original post I advised the OP to use "Grounded Thoery" based on the fact that she has chosen qual data and interviewing as it is a good methodolgoy to fit around already done qual research, to which a person in her class responded that they already had to submit their methododlogy in their proposal- to which I replied she should maybe stick to her already submitted proposal then.


I was responding to you in regards to the porposal (which incidentally DCU states should contain the persons research methodology on their website).

Sinall

PK2008 wrote: »

Im not geting in an argument with you but if you read my original post I advised the OP to use "Grounded Thoery" based on the fact that she has chosen qual data and interviewing as it is a good methodolgoy to fit around already done qual research, to which a person in her class responded that they already had to submit their methododlogy in their proposal- to which I replied she should maybe stick to her already submitted proposal then.


I was responding to you in regards to the porposal (which incidentally DCU states should contain the persons research methodology on their website).

I'm sure, having recently done a thesis yourself, you know that the more information you get and the more research you do on it, the more qualified you are to make decisions on it! My thesis proposal did contain a draft methodology section, but it was very, very small and DCU give you some kind of a template for this. I know, as time went by and I became more knowledgable about the topic, I did revise my original opinion.

I'm not trying to argue with you either - and I know we are both trying to help the OP! But in DCU, the thesis proposal really was quite short and it was mostly for the purposes of assigning you to a appropriate supervisor (one who had an academic interest in that area).

Rowley Birkin QC

Finished mine about 10 weeks ago and went through the same stresses, to be honest phone calls and personal meetings are your friends, e-mail is too haphazard when time is running short.

Above all else, don't panic. You will get it done.

AARRRGH

You're going to be ok.

I did a masters last year at night (work during the day) and picked a horribly, horribly complex topic for my thesis. I was stressed to hell as it was just so difficult.

Here is the thread I made: http://www.boards.ie/vbulletin/showthread.php?t=2055306356

Have a read of it.

The moral of the story is if you work your ass off and stay in touch with your tutor, you'll be ok. Make sure your tutor can see how hard you are working so he/she will recommend you get a fair grade.

G86

Hi guys

Thanks for the replies. Just to clarify: I know what methology I'm using and I've written about this throughout my lit review and also justified it's usage in my methodology section. The problem that I have is that I'm unsure of how to analyse the data I collect. I'm doing semi structured interviews. I did get a mail a poster on here though which was really helpful and suggested coding, which I think I'm going to go with. I did look into grounded theory but I don't think I'd opt for that one.

The main problem I'm having is that my findings are going to be extremely vague and I'm not sure how to nail down the points in reference to my thesis topic. Hopefully coding and thus identifying running themes will help me to do that.

My tutor is unavailable over the summer so personal meetings or email exchanges with her aren't an option. I did email her and she said she was sorry but that they were not allowed to advise students or offer advice during the summer months.

Thanks for all of the help guys, I really appreciate it, it just all got on top of me earlier. Hopefully I'll get to grips with it.

G86

PK2008 wrote: »

Ok I dont know how DCU work it but you should have had to submit your research methodology before you started your research otherwise how can you be sure you are approaching it in the right way??

It sounds like you've done some data collection and are now trying to fit a methodology around it, if so, then i assume you havent formulated a hypothesis yet.

For qualititate research where a hypothesis hasnt been formulated (exploratory) then "Grounded theory" is a apopular methodology.

Look it up but basically it is is a methodology which starts out by collecting qualitaitve data then coding it, then formulating a hypothesis based on the recurring "themes" in the data.

I haven't done any data collection yet, I've simply done a lit review and a methodology proposal. I'm using the methodology from that proposal, the problem is that I'm unsure of how to interpret the data retrieved from it. I got a mark of 78 for my lit review/methodology and I was happy enough with that so I'm sticking with that method. I looked up Grounded Theory but having learned nothing about it in class I really don't want to spend too much of the little time I have left on it now, you know? Also, I already have a hypothesis so I'm guessing it wouldn't be viable.

Sorry if my post wasn't very clear:o

PK2008

G86 wrote: »

I haven't done any data collection yet, I've simply done a lit review and a methodology proposal. I'm using the methodology from that proposal, the problem is that I'm unsure of how to interpret the data retrieved from it. I got a mark of 78 for my lit review/methodology and I was happy enough with that so I'm sticking with that method. I looked up Grounded Theory but having learned nothing about it in class I really don't want to spend too much of the little time I have left on it now, you know? Also, I already have a hypothesis so I'm guessing it wouldn't be viable.

Sorry if my post wasn't very clear:o

Yeh, I know what you mean.

Coding is a good way to go and its pretty easy- theres a few different methods (open, selective, axial etc)- they generally look at extracting a quant measure from qual information.

For example if you asked a question which sought an interviewee's opinion, which you could define as either negative or positive (or some other measure) and they gave you an opinion which matched your "positive" criteria- then you had 8/10 others who also gave psoitive opinions then you could code that as an 80% positive response rate.

You can also code by "theme" for example if you asked the interviewee's to name something that gives them "the biggest problem in their company" and alot of them responded with problems that were similar in nature or symptomatic of a general issue then you can code that as being representative.

Coding is a popular qual interpretive technique (its also part of grounded theory btw) and while I know you dont have time to read new material I think you may need to read up on it a bit so you can describe it in your thesis.

I wouldnt use content analysis unless you have already put together a dataset (or have access to one)- they can be highly interpretive anyway so unless your good with stats you could get caught out.

G86

Thanks a mil, I'm thinking on going for themes, now I just have to figure out what they are..:)

I'm terrible with stats, hence the preference for Qual

PK2008

G86 wrote: »

Thanks a mil, I'm thinking on going for themes, now I just have to figure out what they are..:)

I'm terrible with stats, hence the preference for Qual

Ok well if you're coding themes then that basically is "Grounded theory"

It gets its name because a theme has recurred so many times that you come up with a theory (ie your argument/hypothesis etc) which is "grounded" in your data.

Becuase its a theme its subjective- it may not even be phrased the same way by different interviewee's, but becuse you've identified the running theme you are able to "ground" it (ie show that your not just pulling it out of the air)

If you're going with themes then you need to say that you did your interviews in "rounds".

Rounds are used to identify themes- for example if you had 20 interviewee's you might do 5 rounds of 4 interviews- after each round you try to identify a theme, then you might change your interview style/questions etc for the next round in order to develop the theme. By the end of the rounds you should be able to show how you idenitifed the theme (in which round it became apparant) and how you homed in on it and developed it until such a point where you were confident it could form the basis of your argument.

Themes always remain subjective so dont worry about absolute proof or standard deviations as in quant- all you're trying to do is build up a strong subjective case. ie your study could have started out asking "what is the biggest problem facing companies today" and from responses you ended up with a running theme of redundancies effecting retained workers morale/productivity ie "20/25 of my interviewee's felt that after redundancies the remaining workers morale was very low and this effected their productivity"

At this point you may form a hypothesis for example:
Hypothesis 1= Redundancies effect retained staff morale
H2 = Poor morale has an effect on productivity
and so on-

Obviously the hypotheses relate to your initial research question- but the main point here is that you retroactively arrived at your hypothesis- ie you had to go through your data collection and analysis stages first in order to arrive at your hypothesis

EDIT: i just saw that you said you have already stated your hypthesis, so maybe grounded theory wouldnt work for you, however you could probably stop after the coding stage and just use the coded themes- just make sure they tie into your original hypothesis (this may mean directing your interviewee's toward your hypthosis (which may not be ethical but with your limited time you may not have any choice)

G86

I'm confused now. I have a hypotheses,so I'm not doing grounded theory, I don't want to change my hypothesis. From what I've read/been told, I thought coding was different to it. I thought I could pick out quote, establish themes for each, and work through it that way.

G86

20 interviewees you say???

..I have 3. This just gets better and better..
The thing is though that I wrote that in my proposed methodology and she said it was fine

Sinall

G86 wrote: »

20 interviewees you say???

..I have 3. This just gets better and better..
The thing is though that I wrote that in my proposed methodology and she said it was fine

It depends on your research topic. I had 81 respondents to an online survey and did 4 semi-structured interviews. Another girl in my class did 2 really indepth interviews, another one did about 5. Someone else did solely data analysis on about 200 answers to a survey.

If your supervisor said 3 is fine then it is fine. One of the theses that I brought in to send you (you should get it in a few mins) did about 5 or 6 interviews.

PK2008

Yeh it depends on your research topic- I mean if 3 people constitutes enough of a representative sample then 3 is all you need.

If you've already stated your hypothesis then I assume the questions you asked (the data you collected) were based around testing the hypothesis, ie the 3 people you interviewed were able to give definitive authoritative answers to your hypothesis.

As long as you have justified them as being a legitimate sample then just use their direct answers as evidence (backed up with a bit of secondary research if possible)

If you've already got your hypothesis then theres no need to identify running themes- all you need to do is identify the response rate to your hypothesis questions; For example H1= "The weather effects peoples moods" - 2 respondants agreed, 1 disagreed= 66% agree, 33% disagree

G86

I don't have a means of secondary research so I think I might just stick with the interviews but do 2 more also. I've been sent some brilliant information from posting here that's going to be a major help to me and I really appreciate it I also got the layout out for the thesis which has been really helpful as I didn't have it and it makes things so much clearer.

I'm going to go with transcript analysis to interpret my findings and incorporate the use of coding into that.

Thanks a million for all of your help guys, I really do appreciate it, my head is alot clearer today thanks to the info some of you have mailed on to me and the people who have posted.

I'm going to kick this thesis's ass

dumbblonde122

Sinall wrote: »

It depends on your research topic. I had 81 respondents to an online survey and did 4 semi-structured interviews. Another girl in my class did 2 really indepth interviews, another one did about 5. Someone else did solely data analysis on about 200 answers to a survey.

If your supervisor said 3 is fine then it is fine. One of the theses that I brought in to send you (you should get it in a few mins) did about 5 or 6 interviews.

How would I go about setting up an online survey??

AARRRGH

dumbblonde122 wrote: »

How would I go about setting up an online survey??

http://www.surveymonkey.com/

Sinall

AARRRGH wrote: »

http://www.surveymonkey.com/

Yep, that's the one I used. They have a 24 hour support which is great for dealing with any queries.

The basic account only allows you to make a questionnaire with 10 questions, but I upgraded to a professional account (it was only $19) and then could do whatever I needed to with it!

Surveymonkey is also pretty handy at letting you filter and crosstabulate responses.