How we change what others think, feel, believe and do

A general research framework

Explanations > Social Research > Design > A general research framework

Problem | Question | Hypothesis | Design | Data | Analysis | Conclusion | See also

There are some common elements to research projects which can be used to shape the whole activity. Here's a general approach.

Problem

A research study starts with a problem that they have determined within their broad area of interest. This could be along the lines of:

• Indication of something hidden that might be revealed by study
• The occurrence of an event for which there is no adequate explanation
• An apparent relationship between items which is not explained by current theories

Deep research requires a lot of work and patience, including backtracking after exploration up blind alleys. This usually requires that the researcher is strongly motivated to stay the course.

The problem must also at least appear to be soluble by the researcher, in the time available, with the skills and resources they have. It is no good trying to find the secret of life, the universe and everything if you do not have a lot of time, money and brains.

Question

The next stage is to formulate the research question. This creates focus and may require a significant thinning down of the original problem.

If the problem starts out as being about life, the universe and everything, then perhaps the research question should realistically be constrained to furthering understanding of some aspect of life, such as determining the first stirrings of a fruit fly's egg or the value put on life by people of different ages.

Hypothesis

Developing the hypothesis involves converting the question into a predictive form and also creating a null hypothesis by which falsification may be achieved.

Design

Design of the experiment can be a critical stage as an incorrect design will produce invalid and useless data from which false conclusions may be drawn.

The purpose of the design is thus to determine a method which creates accurate and unbiased data from which valid conclusions may be drawn. This includes determining includes how experimental closure will be achieved.

A big problem in social study is that when people know they are being observed they tend to act differently. This requires careful design to eliminate this bias.

An important part of experimental design is to ensure that all variables other than those of interest are held stable and do not distort the results. One way of doing this is to include a control group, in which the experiment is repeated under the same conditions but without manipulating the independent variable. The results of the two studies may then be compared with the assumption that differences are due only to manipulation of the independent variable.

Where traditional experimental control and management of variables is not possible or not desirable, other methods such as surveys, interviews or more distant observation may be used.

Data

Data design not only includes identification of what data is needed -- it also includes design of how the data will be collected. Measurement of data typically involves manipulating independent variables and measuring dependent variables.

Data may also be gained by observation of naturally occurring events. In such situations the researcher will try not to let their observation affect the data. Two opposing ways of doing this is first to be so separated from the people being studied that you are not noticed (such as using one-way mirrors or hidden cameras). Secondly, you can be so obviously present that people eventually ignore you and revert to natural behavior (such as in reality TV shows).

Choosing the data you will gather has a very significant effect on the analysis and conclusions you will be able to draw. If you want significant and credible results, then data design is a critical activity.

Gathering data is often the most time-consuming and expensive part of the experiment. Designing data to collect thus needs a pragmatic approach that will enable you to conclude useful results without breaking the bank or taking forever.

Where everyone cannot be accessed, careful sampling is used to enable accurate analysis and valid results.

Analysis

After gathering of data, the next stage is to analyze it, effectively turning data into useful information.

Where there is sufficient data, statistical analysis may be used, where tools such as SPSS and SAS may help (although a simple spreadsheet may also be adequate).

Analysis can be quite sophisticated and there are many tripwires where information derived is not as significant as might be supposed.

Considerations about analysis should not be left until after the data has been collected. Deciding what analysis you will do is a part of the design process. This also includes consideration of what conclusions you may wish to draw.

Conclusion

Finally, the analysis is reviewed and specific conclusions drawn that relate to the original question and hypothesis. This can include conclusion about:

• The existence or not of something.
• Correlation between two variables.
• Causal relationships between two or more events.
• The certainty which may be attached to these conclusions.

See also

Research frameworks