Jeromy Anglim's Blog: Psychology and Statistics

Friday, December 11, 2009

How to Write the Introduction Section in Psychology: Theses, Lab Reports, and Journal Articles

This post discusses how to write an effective introduction in a psychological empirical report, including lab reports, theses, and journal articles. It discusses: (1) the overall purpose of the introduction; (2) the different elements of the introduction, including the opening, literature review, and the current study; (3) issues associated with main and transitional introductions in multiple study investigations; and (4) provides links to additional resources on writing introductions.

OVERVIEW

The APA Style manual states that when reporting an empirical investigation the purpose of the introduction is to "[develop] ... the problem under investigation and [state] ... the purpose of the investigation" (p.7 5th Edition).

When I think about the introduction, I often apply what I call The Justification Formula.

Justification = Gap * Importance * Method

Put simply, research is justified by a gap in the literature, where it is important that the gap be filled, and the proposed method goes some way to closing that gap. This formula links to the aims of the introduction to:
1. Introduce the research context;
2. Identify a gap in existing knowledge;
3. State why resolving the gap is important;
4. Set out how the proposed study helps to close the gap.

The Gap

Disagreement in the literature is one way to identify a gap. Some common types of gaps include:
• Disagreement between a theory and empirical findings
• Disagreement between two theories, models, schools of thoughts, or authors
• Methodological flaws in existing empirical research
• The absence of research on a particular topic (e.g., the first to apply a useful method to a topic)
• Inadequate amount of empirical research on a topic
Describing a gap often requires you to identify what the literature "believes". There are several reasons why this can be challenging. First, the articles that make up the literature may express different views. Second, articles may not directly comment on your issue.

Several strategies for dealing with this challenge exist. First, the writing in recent and influential articles represents important evidence of the position of the literature on a topic. That said, if researchers have expressed differing views, these need to be characterised. When characterising a gap based on a beliefs of the literature, even if some researchers agree with your position, the presence of many researchers who don't, does create a form of gap.

In other situations you may wish to define a gap based on an absence of research on a topic. This is also challenging, especially if the topic is cross-disciplinary. The usual strategies of searching the literature can help to reassure you that such research does not exist. Checking reviews and the references of important recent articles in the area is a further check. Forward citation strategies available in some databases are also useful. Depending on the topic, it may be appropriate to state that "to the best of my knowledge, no research has looked at this".

Importance:

Beyond showing that there is a gap in current knowledge, it is useful to show why filling the gap is important. Importance can be justified in many ways. First, it can be justified in terms of the societal benefits. Societal benefit of this sort are particularly easy to justify if you are doing clearly applied research. For example, if you are developing a clinical intervention, an educational program, of an improved selection and recruitment test, the societal benefits that would flow from a successful project are clear. If the research is more pure, other arguments may be used: the process or topic itself may form the basis of applied research. For example, better understanding of cognitive processes is relevant to a wide range human endeavours from education to human factors to artificial intelligence and more.

Second, the importance of a topics can be justified by showing how many people in the discipline are concerned with the topic. This does raise the question of why the discipline is concerned with the topic. Thus, importance can be justified by direct statements about applied benefits or by more general benefits for knowledge.

There is also a distinction between the importance of closing the gap and the importance of the contribution that the proposed study will make to proposing the gap. For example, recognising the importance of developing better measures of emotional intelligence is useful, but this only creates potential importance. For the investigation to actually be important, it also needs to adopt a plan which offers a reasonable opportunity of actually improving measurement of emotional intelligence.

Some specific strategies for justifying importance include to:
• Show that influential authors in the discipline have felt the topic was important
• Show that many authors in the discipline have studied the topic. This can be supplemented with a basic bibliometric count of the number of articles on the topic in recent years.
• Present statistics that show something about the scale of the problem: e.g., numbers of people affected or the amount of money spent each year on the issue.

Method:

The introduction needs to propose a method for closing the gap. The proposed method should be appropriate for the gap. "Method" in the context of the introduction refers to a sufficient description of the strategy to show the relevance of the proposed method.

The following provide some examples:
• PROBLEM: no research on the topic SOLUTION: empirical research is required
• PROBLEM: Existing research has methodological problems SOLUTION: Research is required that overcomes the methodological problems
• PROBLEM: much research, but inconsistent findings or a lack of synthesis; SOLUTION: a meta-analysis may be appropriate
• PROBLEM: poor or conflicting theories; SOLUTION: empirical study to test competing predictions or theoretical development
• PROBLEM: inadequate statistical analyses; SOLUTION: re-analyse existing data or collect new data and analyse
Justifying the method requires an appropriate mapping between problem and solution--gap and method.

SECTIONS OF AN INTRODUCTION

The APA Style Manual states that the introduction should: (1) introduce the problem, (2) develop the background, and (3) state the purpose and rationale. Kendall, Silk and Chu (2000) state similarly that an introduction is made up of three elements: (1) Opening, (2) Literature Review, and (3) Transition to Your Study. I'll use Kendall et al's framework to organise points about each of these elements.

Opening

The opening sets out: the research question; the importance of research question; the gap in current understanding; and the aims of the research.

The opening paragraph in particular typically aims to generate interest, using non-technical language. Kendall, Silk and Chu (2000) mention six specific strategies for the opening paragraph: (1) rhetorical questions, (2) everyday experience, (3) analogy/metaphor, (4) striking statistic/fact, (5) historical fact, (6) lack of previous research.

The opening often includes citations to influential researchers and schools of thoughts. It also sometimes throws in competing empirical findings and theoretical perspectives.

The opening is quite short. One to four paragraphs is common. While format varies, the opening section typically moves from showing why the research is important, to situating the research in context, and then to setting an outline of what is to follow.

Literature Review

I previously posted an extensive guide on how to write a literature review This previous post was written with a dedicated literature review as the prototypical case. However, the ideas can be readily applied to the literature review component of the introduction of an empirical report. Discussion here focuses specifically on the sequential aspects of writing a literature review relevant to justifying an empirical study.

The literature review typically involves a selective historical presentation of the relevant literature. It generally transitions from general to specific issues. It links the historical context to the current research. It summarises what is known and not known and what is agreed and not agreed. The following discusses some specific issues.
Article selection:
When the literature is huge, it is a challenge to decide which studies to include.

Seminal authors and studies often provide a good starting point. The first study to post a question, develop a methodology, develop a theory, or present empirical data on a question can be a good starting point for the literature review.
Detailed discussion of a past study:
Some literature reviews discuss a small number of previous studies in depth. This is a useful strategy when the past studies represent current practice or knowledge on a topic. The identification of flaws or gaps in these studies can function as a launching pad for your study.

A common sequence for detailed discussion is as follows:
1. Justify detailed discussion of specific study: Show that the study is influential and relevant to your study.
2. Describe Specific Study: Set out aim, method, results, and conclusions in sufficient detail to introduce your critique.
3. Critique: Present critique, potentially in separate paragraphs,
4. Draw implications: Summarise what the preceding discussion means for the present research.
Critical Analysis
A literature review should represent a critical reading. In particular when justifying a gap in the literature, a critical perspective may be required to highlight inadequacies of previous research. This should be focused so that it leads naturally to a justification for your study. As always it is important to be respectful of past authors when critiquing the literature.
If the literature review is long, linking paragraphs or sentences are often included. These are typically presented at the end of sections and serve to show where the line of argument is going and what will come next. Headings and the conventions of empirical reports can help to reduce the need for such linking paragraphs. Linking paragraphs typically involve one or more of the following elements: (a) a summary of the preceding section; (b) a link between the preceding section and the aims of the study; and (c) a link to the next section.

The Current Study

This section functions as a transition from the Introduction to the Method. Some common headings for this section include: "The Current Investigation" and "The Present Study". The section is typically fairly brief with one to four paragraphs being common. It typically summarises the study design, links the design with the aims, and states and justifies the expected outcomes.
Design
A summary of the design of the study should be provided. This typically includes an operational definition of key variables. Such variables include any variables experimentally manipulated and major dependent variables.
A justification should be provided for how the design helps to achieves the aims of the investigation.
Expectations
Expectations about the results should be stated. The reasons for these expectations should also be specified. In some instances these are expressed in an enumerated list of hypotheses. In other cases, expectations are expressed more loosely. I have a separate post on how to write predictions and hypotheses.

MAIN AND TRANSITIONAL INTRODUCTIONS IN MULTIPLE STUDY INVESTIGATIONS

The above section focused on the writing requirements of an introduction to single study empirical paper. This section looks at how things change when writing a multiple study investigation. This includes multiple study theses and journal articles. In particular a multiple study investigation has two types of introductions. The first type is what I call the main introduction. This opens the paper and includes all the elements of a single study introduction except that the "Current Investigation" is replaced by an overview of the "Investigations". The second type is the transitional introduction. Before presenting the method and results for each study, a transitional introduction is typically presented.

Main Introduction

The main introduction of a multiple study investigation is typically similar to a single study introduction. The main difference is that instead of having a section on the "Current Study", it has a section that gives an overview of the "Current Studies". This section might be called: "The Present/Current Studies/Experiments" or "Overview of Experiments/Studies".

This section may include:
• Number: The Number of experiments conducted
• Commonalities: Common features regarding aims, design, and hypotheses across the studies
• Differences: An outline of how the studies differ regarding aims, design, and hypothesis
• Sequential Logic: The logic for the sequence including sequential development of ideas and systematic exploration.

Transitional Introduction

The length of a transitional introduction varies substantially. In particular, journal articles often place a premium on being concise. In contrasts theses often have more words to work with. Theses also often require greater justification of decisions. Thus, The following discusses what might be included in a comprehensive transitional introduction, while noting that in some instances, particularly journal articles the transitional introduction may be as brief as a single paragraph.

Various formats exist for presenting a multi-study investigations. I wish to focus on format where there is: (1) a main introduction; (2) a series of studies presented one at a time; (3) a main discussion. In this format each study typically has its own transitional introduction.

Purpose: The purpose of transitional introductions is to introduce and justify the next study. Much of the justification will typically flow from the main introduction. But the transitional introduction serves to reiterate the aims of the study and link the studies in a logical way.

Sequential Development The presence of a sequential development of studies leads to particular requirements for transitional introductions. I use the term sequential development to refer to the way that a previous study influences a subsequent study. The conduct or analysis of a preceding study may have highlighted methodological problems or new and interesting hypotheses. In other cases, a thesis may have multiple studies with no sequential development, or the nature of the sequence was choreographed from the start. The nature of sequential development needs to be set out in the transitional introduction. If we assume a two study investigation some common examples include:
• Study 1 obtains a result, but there are multiple interpretations of the results; Study 2 modifies the design in order to rule out one or more alternative interpretations
• The conduct of Study 1 revealed a flaw in the design (e.g., inadequate sample size; participants were cheating; the system crashed; unreliable measurement; the list goes on); Study 2 avoids this flaw

The three elements The three elements of opening, literature review, and current study also provides a way of framing discussion of transitional introductions.

Opening: This section reiterates what has been learnt so far and set out what remains to be known. It should set up a motivation for the subsequent study.

Literature Review A literature review is generally not required. The main introduction will have covered most issues. The closest thing to a literature review is typically a critical description of the preceding study.

The Current Study This section differs the least from transitional to single study introduction. Design, links between designs and aims, and hypotheses are often set out. For studies based on sequential development, this is often framed more in terms of how the current study differs from preceding studies. However, some studies prefer to put most of this information in the main introduction.

Amount of content to include Several factors will influence the amount of content to include in a transitional introduction:
• Journal article or Thesis: Journal articles will tend to be briefer.
• Word count limitations: Word count limitations will dictate the depth. This partly explains differences in depth between journal articles and theses.
• Amount needing explaining: The more a second and subsequent differs from the previous or the more it attempts to critique or build on the preceding study, the more information will be needed to explain this.

RESOURCES

Offline Materials

• Philip C. Kendall, Jennifer S. Silk, and Brian C. Chu (2000). Introducing Your Research Report: Writing the Introduction, in Guide to Publishing in Psychology Journals.
• APA Style Manual: For the 5th Edition, see section 1.08. Section 6.02 has a few comments about student theses and lab reports.