go to the School of Psychology home page Go to the UNE home page
Chapter 7 - Analysing the Data Part IV - Analysis of Variance Chapter 1 - Behavioural Science and research Chapter 2 - Research Design Chapter 3 - Collecting the Data Chapter 4 - Analysing the Data Part I - Descriptive Statistics Chapter 5 - Analysing the Data Part II - Inferential Statistics Chapter 6 - Analysing the Data Part III - Common Statistical Tests The Scientific Approach The Research Process Variables Confounds The Logic of Cause and Effect Research Hypothesis and Predicition Readings and Links

Chapter 1
Behavioural Science and Research


The Research Process

The "impossibility of proof" philosophy that underlies psychological research illustrates how research is a process rather than a linear event with a clear beginning and end. One study feeds into the literature that is used to spark further inquiry into the phenomenon. As more and more studies accumulate, we gain a deeper understanding of the phenomenon under investigation. But as our knowledge deepens, so to do the questions and ambiguities. Eventually, we appreciate that each single study is a small piece of information in a large and complex puzzle.

Even within a single research project, there is only a rough linear sequence of events in the research process. While we can probably delineate a clear beginning and end to the process, the various steps within it are interconnected. Below IÕve presented a possible model (Figure 1.1) for how the research process unfolds within a single research project. In this diagram, the starting points in the process are indicated in large bold text. The linkages in the process are depicted with arrows, with double headed arrows meaning that process "flows" in both directions. There are also two "feedback" loops in this diagram, depicted as dashed lines that indicate that earlier parts of the process may be revisited temporarily at later stages.

One possible model for the research process

Figure 1.1. One possible model of the research process.

Research is based on the empirical testing of ideas. These ideas might come from any number of sources. They might start with a theory about some phenomenon. That theory might rely on the literature in the field and together the theory and the literature are used to devise a study. Or a researcher might notice in the literature that a series of findings are conceptually related. From these observations the psychologist might devise a theory to explain the findings. Or, more simply, the researcher might just be curious. Perhaps sheÕs seen a particular behaviour occur in a particular circumstance, and wonders why. To answer this question, the researcher thinks about the problem, looks at relevant studies in the literature, and tests her explanation by conducting a study. The researcher needs to focus the question(s) of interest into specific testable research hypotheses.

Study design is a major part of the whole process. Unless the data are gathered in a systematic and unbiased way, no amount of statistical conjuring can retrieve the situation (i.e., you need to start with quality information). There are many ways of designing studies but the main aim is to answer as clealry and unequivocally as possible the particular research hypotehses that have been pose. What is the best way of gathering the best data possible given the constraints of economics and time? We will be looking at many major design features.

After the data are collected, data analysis proceeds. The data analysis would first focus on the specific research hypotheses, to see if the hypotheses or predictions are supported by the data. A good data analyst, however, will also explore the data to see if anything unexpected or unanticipated emerges. Some of these analyses may be guided by the literature in relevant studies. Or the investigator may find something unexpected and, in an attempt to further explore these unexpected findings, looks to the literature for some guidance. The researcher may also discover at the data analysis phase that there are unanticipated alternative explanations for a finding or set of findings. A new study might be devised to examine these alternative explanations.

After the data analysis is complete, the results are written up in report form. Research reports are often, though not always, written up using what is called APA format (APA stands for American Psychological Association). This is simply a standardised means for formatting a research report so that it flows in a manner that a reader will anticipate (the sections, broadly, are the Abstract, Introduction, Method, Results, Discussion, and References). The format the investigator uses will depend on the research journal that the investigator is targeting for publication (such as the Journal of Personality and Social Psychology, or Learning and Memory, or the Journal of Clinical Psychology). Not all journals use APA format, but many use APA or some variant of APA format. Notice in the diagram a feedback link between the writing of the report and the literature. In an empirical research report, the writer is expected to make the link between the current study and the existing literature very explicit. One criterion for publishing a research report that journal editors often use is whether or not the research is sensitive and relevant to the current research literature.

Once the report has been written, it is submitted to the editor of the scientific journal, who decides first whether or not the manuscript is appropriate for the journal. If so, the editor will typically then send the manuscript to two or more experts in the field who have the knowledge and expertise to critically evaluate the research (this is called peer review.) A reviewer will be looking to see if the research described in the manuscript is well conducted, important, interesting, and ultimately useful for the rest of the research community. Manuscripts are either rejected outright, or the reviewers will suggest a number of modifications to the paper, more data be collected, etc. In such a case, the editor would give the original investigator the comments of the reviewers with a request that the manuscript be "revised and resubmitted". In rare cases, a manuscript will be accepted without revision.

The revision process may be as simple as clarifying a few ambiguities, rephrasing key points, or addressing some relevant literature in the paper. Or the revision may be quite difficult, requiring extra data collection, additional data analyses, or rewriting of entire sections of the manuscript. Almost all research papers, if they are not rejected (most are), require some revision before publication. Once the revisions are completed, the paper is resubmitted. The editor will then send the paper back to the original reviewers, or perhaps a couple of new ones. This process continues until the editor decides to publish the paper, or finally reject it. If accepted, the paper becomes part of the literature for others to examine, and the cycle continues.

In this class, most of our time will be devoted to the "study design," "data collection," and "data analysis" components of the research process.




© Copyright 2000 University of New England, Armidale, NSW, 2351. All rights reserved

UNE homepage Maintained by Dr Ian Price
Email: iprice@turing.une.edu.au