Developing research ideas and formulating research questions

Chris Brunsdon

1

Research Projects in general

Most research projects share the same general structure

Broad area of interest may be too broad to study in any single research project

From

Hourglass

You have done some background research

Reading the literature around your topic

Talking to supervisors

The aim is narrow down your Question

Formulate a hypothesis or research question

Then for the actual dissertation (not the proposal)

You will be engaged in direct measurement or observation of the question of interest

At the narrowest point

After basic data is collected you will

Begin to try to understand it

Perform analysis

After analysis you will

Start to formulate some initial conclusions

Generalize from the results of your study to other work

Hourglass

Background research

Narrow down your research question

Direct measurement

Analysis after basic data is collected

Reflections

Research Question

Initial research ideas are often very general

But good research questions need to be quite specific, so that:
– appropriate methods can be chosen
– required resources can be identified
– criteria for success are clear
– work can be planned realistically
– useful objectives can be met

Research ideas

General ideas may flow from a number of basic aims

Professional – ‘I want to develop a new spatial analysis technique’

Methodological – ‘I want to use R-based image processing software’

Ethical – ‘I want do something that makes a difference to the environment’

Turning ideas into questions

A good research question is clear and specific

Answerable using data / resources available

Consider the subject matter:

who, what & why?

Consider the practical:

where, when & how?

Then review the question to see if it really requires research

The subject matter

‘Who’ are the ‘subjects’?

General: land cover change

Specific: tropical rainforest

‘What’ is the test?

General: remote sensing

Specific: fuzzy classification vs. other methods

‘Why’ considers is this a good thing to do?

General: to manage remote sensing uncertainty

Specific: to improve monitoring of specific forest resources

Practical issues

‘Where’ is it possible to do this?

General: Tropical rainforest areas, Data

Specific: Bolivian jungle

‘When’ – is there time to do this analysis?

General: this year

Specific: complete research by August 12th

‘How’ can the plans be implemented?

General: I have the necessary skills

Specific: I will have learn to use new RS software

Research question

If the general idea was to analyse land cover change, then the specific question might be:
‘Does fuzzy classification of remotely sensed imagery improve land cover change monitoring in tropical rainforest environments?’

It should be clear from thinking about the practical issues that it not realistic to answer this question robustly using an analysis of just one rainforest area!

Research Question

Your final research question should be realistic

The methods should be appropriate to meet the stated objectives

‘Comparison of Fuzzy and Boolean classification methods for analysing tropical rainforest change: a Bolivian case study’

Evaluate Your Own Research Question

Is the topic interesting to you?

Does the question deal with a topic or issue that interests me enough to spark my own thoughts and opinions?

Is the question easily and fully researchable?

What type of data do I need to answer the research question?

The research question, “What impact has the ‘Agenda for Change’ had on the provision of Primary care in Nottingham” will require certain data:

statistics on Primary care before and after (illness, location)

statistics on Primary care providers, (type, location)

information about practice before and after ‘Agenda for Change’ recommendations

Evaluate Your Own Research Question

Is the scope of this information reasonable

Given the above …

…the type and scope of the information needed

is the question too broad, too narrow, or okay?

What sources will have the information that I need to answer the research question?

journals, books, Internet resources, government documents, people

Can I access these sources?

Given my answers to the above questions, do I have a good quality research question that I actually will be able to answer by doing research?

image1.png