1 / 15

Writing up results

Writing up results. Results are divided into two main sections , usually. Descriptive Statistics Include frequencies for nominal/categorical variables Include means and standard D for continuous/ranked variables Include # of missing (not items, but actual surveys that could not be included ).

cera
Download Presentation

Writing up results

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Writing up results

  2. Results are divided into two main sections, usually • Descriptive Statistics • Include frequencies for nominal/categorical variables • Include means and standard D for continuous/ranked variables • Include # of missing (not items, but actual surveys that could not be included)

  3. Some uncertainties • Descriptive statistics (means, sd) are often compared to similar published data to see if they make sense. • E.g. “CES-D scores had a mean of 8, SD of .79 in the current sample. This is similar to a same aged sample collected by Smith and Jones (2011). • This kind of info can go in results or sometimes in discussion. I prefer it in results. • You don’t have to include it, but if you do that is where I would like it.

  4. Where do the alphas go? • Alpha tells us if folks answered questions consistently within the same scale • The alphas go in the method section • Technically they are a “result,” but they go in the method section at the end of the description of the measure.

  5. How do you calculate an alpha? • In SPSS • Go to Analyze/Scale/Reliability Analysis • Select the items in the scale (use the reverse scored items where relevant) • Analyze • Results show if your scale has good internal consistency in this sample or not

  6. Results—substantive analyses • If you have hypotheses, you can call this section “hypothesis testing” • Otherwise you can call it “Research Questions” • Looking at how other published articles do this is the simplest, fastest way to learn how to present these numbers.

  7. Results Section General Framework • Results for each hypothesis can be divided into 3 general statements • 1. What the test was, generally • 2. The actual variables and setup of the analysis • 3. The numerical and substantive outcome of the analysis

  8. 1. mention the type of statistical test you used • Like “We ran correlation analyses to determine if happiness total score was related to the number of reported facebook friends.” • Or “We used analysis of covariance to determine if there were differences in food consumption by experimental or control group, after controlling for body image ratings.”

  9. Results sections • Or • “We used correlation to determine if self-reported frequency of drug use was related to perception of peer risk behavior.”

  10. 2. Specify the variables and setup • This will sometimes seem redundant to the previous statement. If so, you can leave it out or minimize it. • Like “We calculated bivariate correlations between the total score for self reported drug use, and perception of peer drug use.” • This is redundant. You could leave it out.

  11. 2. Specify the variables • It will not always seem redundant. • Like “We entered group membership (experiment or control) as a predictor, and body image rating as a covariate. The dependent variable was ounces of unhealthy snack food consumed.

  12. 2. Specify the variables • Sometimes you will have repetition when you are doing analyses that are parallel. • Like “We conducted a separate ANCOVA with the same predictor and covariate, and the dependent variable for this analysis was ounces of healthy snack food consumed.” • Be sure to specify when the dependent or independent variables change for various analyses, even if they are related

  13. 3. Give the numerical and substantive outcome • The numbers will depend on your test. You need to include the statistic and the significance (p value, usually). • E.g. for a correlation: “self perceived drug use and perception of peer drug use were significantly positively related: r = .14, n=78, p=.04.” • Or “Happiness totals and intensity of facebook use were significantly negative correlated: r=-.20, p=.03.”

  14. …numerical and substantive outcome • If you are presenting ANOVA, ANCOVA or T tests you will also need to include the degrees of freedom. • E.g. “There was a main effect for group membership, such that members of the experimental group were less likely to consume unhealthy snacks: F=3.22 (1, 68), p=.02. Means and standard deviations are presented in Table X.

  15. Numerical and substantive outcome • In a real paper you do not report the numbers for non-findings • In this paper I would like you to report the numbers for non-findings • Also note that statistics such as r, t, F, and p are italicized in apa style.

More Related