Lab 11: How to Write an APA Results and Discussion Andrew Reineberg 11-16-2011.

Lab 11: How to Write an APA Results and Discussion

Andrew Reineberg 11-16-2011

Today we will:

• Learn how to write an APA-styled results and discussion section

APA Results

• For reporting the results of the data analyses used to test your hypotheses.– Identify data analysis technique– Report your test statistics– Put numerical results in word-form– Provide visual representation of data

• Analyses should be related to your hypotheses– If not, state that they are exploratory

General Things To Do

• Start off with general statistics if you have any• For example:

– “Overall, average performance on the memory task was well above chance (M = 93.52, SD = 7.32).”

– “Data for two participants was below chance (50%), thus the analyses reported include data for the full set minus 2 (30 participants).”

1. ID Data Analysis Technique

• General statement about how you tested your hypothesis

• For example: – “A one-way analysis of variance (ANOVA) was conducted to test the

hypothesis that there are gender (male vs. female) differences in the ability to recall single digits.”

– “To investigate the effects of emotionality and gender on face perception, a 2 (positive words vs. negative words) x 2 (male vs. female) ANOVA was performed.”

2. Report Your Test Statistics

• Report:– the descriptive statistics - means (M), standard

deviation(SD)– the test statistic (t value)– degrees of freedom (df = ), – and the probability of the result occurring by

chance (p value)

Formatting

• All decimal places (for p & t values) should be consistent throughout whole paper– usually we use 2 decimal places

• All statistical symbols that are not Greek letters should be italicized – M, SD, t, p, etc.

• When reporting a significant difference between two conditions, indicate the direction of this difference (see next slide)

2 & 3. Examples

• “Results indicated a significant difference in male’s ability to recall single digits versus female’s, t(29) = 1.25, p = .08, such that , on average, men (M = 8.4, SD = 1.3) recalled more digits than females (M = 4.2, SD = 3.1).”

• “Results indicated that the average number of errors was significantly lower in the silence condition (M = 8.11, SD = 4.32) than were those in the white noise condition (M = 12.4, SD = 2.26), F(1, 27) = 8.90, p = .011.”

Notes

• Usually we report exact p values unless they are smaller than .001. – Just say p < .001 if that happens

• If your p values is less than .10 but greater than .05, it is okay to report– There was a marginally significant effect of

gender…

Hypothesis Testing

• Reminder:– p values are important for conveying your results,

but do not overinterpret them• p value is the chance that the result you observed was

due to chance alone, not the probability that your null hypothesis is false• p of .05 is not sacred, but is widely accepted•

4. Visual Representation

• Use simple graphics with easy to read labels and colors/patterns

• DO NOT use 3D graphs– No one can read them

Good Example

Bad Examples

Discussion

• The last content section of a paper• To discuss the interpretation and implications

of your study– evaluate in respect to your hypotheses

• Start off with a brief, non-technical summary of the results. Tell the reader about the main findings without using statistical terminology.

Discussion cont’d

• What do your results mean?– Particularly in light of previous research you highlighted

in introduction• You are also encouraged, when appropriate, to

comment on the importance and relevance of your findings. – How are your findings related to the big picture?

• An ending paragraph in which you make a final summary statement of the conclusions you have drawn.

The Next Few Weeks

• Next week = Fall Break• Nov 30th = Rought Draft Due– Turn in via email by start of your lab section– Workshop for final papers in class

• Last week of class– Final paper due via email by beginning of last class– Q&A session for your class final if you want