How many specific objectives do you have in your doctoral thesis?

How many research hypothesis/ questions do you have?

In terms of methodology:

Do you use cross-sectional or longitudinal design or other types of design?

Do you conduct qualitative analysis? Quantitative analysis? Or combine both? (Mixed methods)

Do you simply adopt or adapt existing scales or maybe develop your own?

Do you use self-rating or multiple ratings in data collection?


I'm just curious to know.

Since nobody replied, let me just share something that is remotely relevant. I believe this applies more to qualitative research.

http://www.timeshighereducation.co.uk/news.../410208.article

1. I have a lot of objectives and research questions.
Eg. To test the relationship between xxx and yyy. (Objectives)
Eg. What is the relationship between xxx and yyy? (Research Question)

2. Yes, I have a lot of hypotheses.

3. Cross-sectional for sure.

4. It's a full quantitative study.

5. Adopt and adapt the measurement for sure. It's not recommended to develop your own scale because it's like reinventing the wheel unless you cannot really find any measurement for the variable you are examining.

6. Self rating, multiple rating???
Do you mean multiple sources vs single source of data?
I will be collecting data from a single source only.

I think it applies to both qualitative and quantitative equally.

Point number 6. in the article - "Assume something you are doing is new because you have not read enough to know that an academic wrote a book on it 20 years ago" is something that always in my mind. Whenever I read thesis or papers claiming that, I'm impressed but at the same time feeling skeptical. I'd think "Maybe the researcher didn't read much/unaware of similar studies.

Do you measure direct relationship? (e.g., A correlates with C), or indirect relationship (e.g., A correlates with C through B)?

Yes, I meant multiple sources of data collection. I will collect data from two sources.

I also will adapt the measurement. Cos there's existing measurement and to develop a new one could take years...

My initial plan is to collect quantitative data, but at last changed to mixed methods.

Yes, I will test for mediating indirect effects.

By right, one shouldn't waste time testing for direct effects from A to C. By including a mediator, one should be able to establish that A is already strongly related to C from past empirical studies. That's the reason why we just need to test the indirect effects.

It would really help the PhD readers, if you could describe the background of your research topic in Social Science.

Science is the collection of grand explanations of things, and scientific research is quantitative in many ways. By quantitative, its thinking relies heavily on linear attributes, measurements, and statistical analysis. But each of the divisions of science also has a qualitative side, in which personal experience, intuition, and skepticism work alongside each other to help refine the theories and experiments. By qualitative, it relies primarily on human perception and understanding.

The purpose of qualitative research is usually not to reach general social science understandings but understandings about a particular situation. By understanding better the complexity of the situation, most social science-oriented researchers can contribute to setting policy and professional practice.

Without knowing the background and the prime objective of your research, the more we answer generically, the more you ask in a roundabout way.

Thanks for the information.

As we know it, there are many levels of phd research. Some are complex and some are simple (Didn't imply the complex contributes more than the simple one). I was just curious how you guys in the social sciences conduct research. The framework, methodology, etc in general. No intention of going into details of how they are going to do it.

Basically my area of research is in hrm. General objective is to link independent variables to dependent variables with a relatively new construct (mediator). Quantitative method only can show the relationships on the surface only. So qualitative method will be conducted to get more understanding of such relationships (like how it works, why it happens in the specific setting, etc).

I usually state three Primary Objectives in my research, then followed by Secondary Objectives, if any. I have three questions for you, which may help you in writing your introduction.

(1) Why is your HRM issue studied important?

(2) How does your study relate to previous research in HRM and how does it differ from other studies on the same issue?

(3) What are the hypotheses and objectives of your study and how do they relate to relevant theory (if they do)?

Adequately testing a theory requires using both confirmational and disconfirmational hypotheses. Usually, you will pursue a confirmational hypothesis when a theory is fresh and relatively untested. The objective during this phase of testing is to determine whether the theory can predict or explain the phenomena within its domain with reasonable precision.

If the theory survives these tests, you will eventually want to pursue a disconfirmational hypothesis. The objective during this phase of testing is to determine whether outcomes that are unexpected from the point of view of the theory nevertheless happen. If unexpected outcomes do occur, it means that the theory is, at best, incomplete. It will have to be developed further so that it can account for the previously unexpected outcome, or it will have to be replaced by a better theory.

An advantage of the Cross-Sectional Design is that it permits you to obtain useful developmental data in a relatively short period of time. You do not have to follow the same participant for 10 years in order to assess age-related changes in behavior. Generation effects are a major problem when you use a cross-sectional design to evaluate age-related changes in behavior of participants of quite disparate ages. The design may be more appropriate when the participants are closer in age.

In Longitudinal Design, a single group of participants is followed over some time period. For example, you could obtain a group of participants and give them cognitive & behavioral tests at 10-year intervals over a 50-year span. In one respect, the longitudinal design circumvents the problem of generation effects that plagues cross-sectional designs. Because you are studying people from the same age group, you need not worry about generational effects when drawing conclusions about that group.

A disadvantage of the cross-sectional and longitudinal designs is their relative inability to determine whether factors other than age are influencing the observed changes in behavior. The Cohort-Sequential Design, described by Schaie (1965), combines the two developmental designs and lets you evaluate the degree of contribution made by factors such as generation effects. However, the cohort-sequential design does not eliminate generation effects. It simply lets you detect them and consider them in interpreting your data.

Thanks for the enlightenment. Good one