Langston, Research Methods Laboratory, Notes 1 -- Method and Results Sections
 
These are the rules for writing method and results sections from the Publication Manual of the American Psychological Association. You should consult the manual if you have additional questions. Where possible, headings, labels, etc. are presented consistent with the rules. Information in {} is descriptive. Information in <> is a variable that you change for your paper. The rest is sample text.
 
Global rules:
1.  Double space the whole paper. Do not put in extra returns between headers and text.
2.  The heading for the first page is "Running head: <RUNNING HEAD> <#>" where <RUNNING HEAD> is a summary of the paper title and <#> is the page number. The title page is page 1, number consecutively after that. The title page and all subsequent pages should have the heading, but all pages after the first page will not have the words "Running head:."
3.  Method and Results sections are in the past tense.
 
Method
{Method sections are used to describe to your reader exactly what you did in your experiment. There are three main parts: Participants (who), Materials (what), and Procedure (how). The general rule for methods sections is to tell enough to make it possible to replicate the experiment, and no more.}
 
Participants
{Participants are the people who participated in your experiment. This will usually require just a couple of sentences. A sample:}
 
     <how many> Middle Tennessee State University students participated in this experiment <why they did it>. There were <how many> male and <how many> female participants. <how many> participants failed to complete the experiment <because>.
 
{Notes:
1. Don't start a sentence with digits, spell out numbers to start sentences.
2. <why they did it> is usually "in partial fulfillment of a course requirement" or "for extra credit."
3. If nobody failed to complete the experiment, leave that part out. Generally, if you have drop-outs, it's useful to know what happened.
4. Depending on the experiment, more information might be required. For example, in a developmental study, the ages of the children would be needed.}
 
Materials
{This is like the ingredients list in a cookbook. List what you needed to perform the experiment. Don't list trivial things like paper and pencil. The best way to start is just to list all of the materials that you used. Free-associate here, don't try to censor yourself. Then, list how each of those items was developed. Once the list is done, turn it into text. Try and describe the materials in some logical order (like order in which they're encountered by the participant). Be sure to include a description of practice materials.}
 
Procedure
{A list of the procedures followed. This is usually a list of actions performed by the participant (like they signed a consent form, read instructions, etc.). To come up with this, take yourself through a sample run of the experiment and make notes on what happens. Generally, you want to structure this in order of occurrence.}
Results
{Present the results of your experiment. This is primarily a vehicle for reporting statistics. Some general rules:
1. Structure it like this:  Tell them what you're going to tell them, tell them, tell them what you told them.
2. Add headings and labels as required to provide structure. If you had a hypothesis, discuss results in the context of the hypothesis.}
 
The data were analyzed using <analysis>.
Recall that the <name> hypothesis predicted <what it predicted>. This hypothesis was/was not supported, <statistics>.
Repeat as necessary.
 
{Samples for each of the major statistics types: (Langston, 2011)}

{Chi-square:}

The stare detection frequency distributions for stare and non-stare trials were compared using a chi-square test of independence. The frequencies are presented in Table 1. The chi-square was not significant, X2(1, N = 80) = 0.09, p > .90. There was not enough evidence to conclude that the two frequency distributions were not independent. Participants were approximately equally likely to “detect” staring whether or not they were being stared at.

{Correlation:}

Magical ideation scores ranged from 26 to 79 (M = 48.6, SD = 9.8). Belief in the paranormal scores ranged from 1.0 to 6.0 (M = 2.7, SD = 1.2). There was a significant, positive correlation between magical ideation and paranormal belief, r = .59, p < .01. Higher magical ideation scores were associated with stronger paranormal belief.

{t test:}

The data were analyzed using a dependent samples t test. The independent variable was the color of the paper over which participants held their hands, and the conditions were red paper and green paper. The dependent variable was rated stickiness on a scale of 1 to 5, with 1 meaning “equal to the baseline stickiness” and 5 meaning “much stickier than baseline.” The mean rated stickiness for red paper was 2.85 (SD = 1.15) and the mean rated stickiness for green paper was 1.70 (SD = 0.73). With alpha = .05, the two population means were significantly different, t(62) = -8.05, p < .01. (Report format from Glenberg, 1996)

{ANOVA
One-way:}

The data were analyzed using a one-way, between-participants ANOVA. The mean for sheep was 9.80 false recalls (SD = 3.04), the mean for neutral participants was 6.55 false recalls (SD = 2.04), and the mean for goats was 3.95 false recalls (SD = 2.11). With alpha = .05, the means were significantly different, F(2, 57) = 28.88, MSE = 5.95, p < .01. Protected t test comparisons indicated that the differences between sheep and neutral participants, sheep and goats, and neutral participants and goats were all significant.

{Factorial:}

The data were analyzed using a two-way, between-participants ANOVA. The factors were story content (arousing, neutral) and presentation direction (forward, backwards). The dependent measure was the participant’s score on an arousal scale. For all analyses, the significance level was set at .05.
If the hypothesis were correct, then we would expect a main effect for story content. In particular, arousal should be higher for the arousing story than the neutral story. This main effect was significant, F(1, 79) = 15.92, MSE = 13.59, p < .01. The means for arousing content and neutral content were 5.82 (SD = 1.7) and 2.90 (SD = 1.2), respectively.
The main effect for presentation direction was also significant, F(1, 79) = 6.77, MSE = 13.59, p = .04. The means for forward and backwards presentations were 4.75 (SD = 1.2) and 4.00 (SD = 1.7), respectively.
The story content by presentation direction interaction was significant, F(1, 79) = 7.73, MSE = 13.59, p = .01. The means are illustrated in Figure 1. Simple main effects analyses indicated that when the text was presented forward, participants reported more arousal for the arousing text than the neutral text. When the text was presented backwards, participants’ reported arousal did not differ.

{References
Glenberg, A. M. (1996). Learning from data: An introduction to statistical reasoning (2nd Ed.). Mahwah, NJ: Erlbaum.}

Research Methods Lab Notes 1
Will Langston

Back to Langston's Research Methods Page