hello.. am in the process of rewriting my thesis under new supervision and this time i want to get things right.. i don't want any room for what happened last time. (for those who remember.)

my question is.. i've been hearing many things about the format of a phd. that the research question/s should be clear etc. so i made sure that mine were clear. then an american academic looked thro it and said that it didn't have a statement of the problem and a hypothesis. what on earth are those?

does anyone has a book with a format for a first chapter or a guideline i can follow to address these questions. i don't want to waffle in my viva trying to say what they are and where they are.

somebody help..

It is usually in the introduction. In some other countries there is sub section as the statement of the problem or the objectives. By the way you should clearly state the problem (or the question). So try to highlight it using the "problem" or "question" as it is the first thing a reader looks at. The objective of your thesis is to solve (tackle) this problem. Hypothesis comes just after the statement of the problem. It is your guess - prediction - about the likely solution of your problem.

true, but what if the question is not a problem but a "what is the effect of...." etc.

you can make hypotheses and sub-questions but can you really call it a problem to solve? or simply expanding the knowledge on a topic?

true, but what if the question is not a problem but a "what is the effect of...." etc.

you can make hypotheses and sub-questions but can you really call it a problem to solve? or simply expanding the knowledge on a topic?

Not all research has to include a hypothesis. It depends on the field and the methodology. For example, if you were doing grounded theory research in the social sciences you wouldn't be expected to have a hypothesis, as the theory is supposed to emerge from the data, and a preconceived hypothesis could influence the findings.

Can you look at some other theses in your field that have passed and see how they have structured the introductory chapter(s)?

Aussiechick is right, it can be common to do 'inductive' rather than 'deductive' research where 'theory' is led by findings. Saying that, in this case you can say that current literature lacks X data which is what you will do and you expect that your work will thus be posited Y here.

Dear Jojo,

As above responders have indicated I think it depends on the kind of research you are doing. The following are some examples without trying to be pedantic.

If a quantitative study, for example a randomised controlled trial, then you are probably doing the study because there is some sort of problem (for example patients with high blood pressure suffer more heart attacks).
You make a hypothesis (reducing blood pressure will reduce heart attacks), and set the research question (does reducing blood pressure reduce heart attacks?)You do the trial: one group with medication, one without (obviously ethically not defendable) and based on the results either confirm (yes reducing blood pressure does reduce...) or reject your hypothesis.

If a qualitative study, there is a problem ( it is unclear why customers buy product X), you set an open question (what are the views of customers on product X), you collect data for example via focus groups and perhaps based on data analysis are able to provide explanations for why people not buy the product.
:-)

Hello everybody,

I am also writing my dissertation. However I got things mixed up a little bit. Generally speaking, my topic is The influence of one thing on another. I stated my research questions and objectives at the beginning. I am doing a quantitative research using questionnaires. However, I was told that quantitative research goes with hypothesis testing. I have no idea how to test hypothesis in SPSS. So my question is can I have only research questions without hypothesis? Is it going to be very wrong if I answer my questions by analyzing the data in SPSS without having to test hypothesis?
Hope I explained it right. thanks

The fact that you think x has an influence on y is in itself a hypothesis and the null hypothesis to be challenged is that x has no influence on y.

Great... and how do I test those two variables when I've got so many variables in my questionnaires?

Well, you must have chosen those variables for some reason surely - you must have had some reason to think they might have some kind of influence on something?

In quantitative analyses, if you don't start with any kind of hypothesis or predictions of any sort, then you usually have to have much more stringent standards (i.e. levels of significance) before interpreting e.g. correlations between variables as being due to any other than the chance of finding anything significant given you are doing a lot of tests (this is generally referred to as 'going on a fishing expedition'). Just something to bear in mind.

Hey Silvia. I'm in a similar position to you for my PhD, and also for my MSc which I finished last year. For the MSc I was told that it was acceptable to have 'research questions' instead of a specific hypothesis, since the research was exploratory. I don't know if the same applies for the PhD thesis, but certainly I have started out with research questions instead of specific hypotheses, because in my field there simply isn't enough existing research to form hypotheses from! Thus I am back to the 'exploratory' type study- but as another poster mentioned, it is important to calculate your minimum sample size accurately to account for the type of analysis and number of calculations you are expecting to perform on the data. Have fun! KB