Survey research and design in psychology/Assessment/Lab report/Feedback/2017
Jump to navigation Jump to search
General feedback about the lab report (2017)
Marking distribution[edit | edit source]
Descriptive statistics (after incorporation of late penalties):
Title[edit | edit source]
- The title page could be improved by using APA style.
- The title should clearly and unambiguously communicate the main content of the report. Weaker titles tend to be more vague, lacking in reference to the main constructs involved in the psychometrics and the relationships investigated in the MLR.
- The Title could more clearly indicate the IVs and DV.
Abstract[edit | edit source]
- The most common area for improvement is to provide less info about the Intro and more info about the Results and Discussion.
- More specifically, what factors were identified in the EFA and how much of the item variance did they explain?
- What was the R2? What was the direction, strength, and significance for each predictor?
- What are the main implications/recommendations?
Introduction[edit | edit source]
- Review research and theory about the factor structure of either time management or time perspective
- Review research and theory about the relation between at least three IVs and a DV
- State a research question to guide the psychometric analysis and logically derived, testable hypotheses for each IV for the MLR
Method[edit | edit source]
Participants[edit | edit source]
- Simple participant profiles focused on age and gender of the sample. Age is skewed, so it is not enough to rely on the mean as an indicator of central tendency. Use frequencies and percentages to describe nominal data.
- More extensive profiles described other sample characteristics from the demographic data available.
- The best profiles also made some comparison between the sample and UC student population statistics.
- Surprisingly few mentioned the proportion of SRD students in the sample who were involved in the data collection.
Measures[edit | edit source]
- Only describe measures for variables included in the data analysis, but provide more detail about how the variables in the current study were measured.
- Provide an APA style citation for the survey and citations for the sources of the items.
- Include a brief summary of the development of the survey instrumentation.
- Type of measurement scale (e.g., Likert?)
- Direction of scoring - were any items reverse-worded?
Procedure[edit | edit source]
- Identify the target population, sampling frame, and sampling technique used.
- Provide an APA style citation to the survey administration guidelines.
- Provide more detail about how the data you contributed was collected (e.g., refusal rate?).
- Comment on any procedural anomalies.
Results[edit | edit source]
Data screening[edit | edit source]
- Only describe screening of data screening used in the current study.
- Provide more detail about the number of cases that were removed and/or why they were removed.
- Provide more detail about how out-of-range values were treated.
- Assumption testing belongs in the section relating to that analysis.
- Avoid mentioning specific case ID numbers - these are not relevant to an unfamiliar reader.
- Avoid mentioning specific variable names in the data file - these are not relevant to an unfamiliar reader.
Psychometrics[edit | edit source]
- It is excessive to examine factorability via any more than one of: correlations, anti-image correlation matrix diagonals, KMO, and Bartlett's test of sphericity - one of these is sufficient.
- PC is recommended over PAF when composite scores are going to be created for use in subsequent analyses.
- Oblimin rotation is recommended over Varimax rotation for correlated factors.
- Present an APA table (not dumped SPSS output) of factor loadings and communalities.
- Sort the factor loadings by size (makes them easier to read).
- Eliminate items which don't satisfy the criteria discussed in Lecture 05 and Tutorial 03.
- Label and describe each factor.
- Composite scores and correlations
- Present a table of descriptive statistics for the composite scores.
- Provide the correlations between the factor composite scores.
Multiple linear regression[edit | edit source]
- Specify the IVs and DV(s) and indicate any recoding or manipulations that have occurred to form these variables e.g., Stress was often used as the DV but with no explanation of how this composite score was created (any items reverse-scored? Cronbach's alpha? Descriptive statistics?)
- After cases which were MVOs were removed, did this have any effect on the Results? If not, retain those cases.
- Present and explain the correlations between items.
- Present and explain the regression coefficients.
- Demonstrate understanding of the strength and direction of relationships.
Discussion[edit | edit source]
- Discussion should consider the factor structure/psychometric properties and hypothesis testing. More discussion was often needed about the factor analysis, factor structure, and other possible factor structures or how to further improve the measure.
- Weaker reports had vague recommendations/implications; stronger reports provided more detail and more positive, practical, and useful applications and recommendations.
- Emphasise the take-home messages.
General comments[edit | edit source]
- Check/correct APA style e.g.,
- Title page
- Running head / page numbers
- Heading style
- Abstract layout
- Title should be repeated at the beginning of the Introduction (and the heading "Introduction" should not be used)
- Citations (e.g., alphabetical ordering of multiple citations, do not include author initials)
- Expression of numbers (e.g,. when used at the beginning of sentences, write as words)
- Table formatting, especially formatting of captions (see Tutorial 01)
- References (e.g., capitalisation, italicisation, do not include the issue number for seriated journals)
- Check/correct grammar, spelling, and proofreading, e.g.,
- Use of punctuation, especially commas
- Use of ownership apostrophes (e.g., student’s vs students' vs students)
- That vs. who
- Use Australian spelling (e.g., hypothesize -> hypothesise)
- when to use "&" vs. "and"
- more proofreading needed
- Quality of written expression:
- Avoid one sentence paragraphs; a paragraph should normally be about one idea expressed in three to five sentences.
- Scientific lab reports should be written in the third person (i.e., avoid 1st person, "we", "our", and "I").
- External assistance (e.g., Study Help) is strongly recommended in order for the quality of written expression to be of professional standard.
- Use page breaks rather than multiple line spaces to separate pages
- Use paragraph styles to indent paragraphs rather than manually tabbing paragraphs or references