OToPS/Measures/template
Appearance
Overview
[edit | edit source]enter text here about measure and uses
History
[edit | edit source]enter text here about general background and history on measure
Main Features
[edit | edit source]Uses
[edit | edit source]- Screening
- Diagnosis
- Service Access
- Monitoring Treatment
- Progress
- Outcome
- Minimally Important Difference or Clinical Significance
- Research
- RDoC connection
Highlights of evolution
[edit | edit source]- Any major version changes
Competitive Comparison
[edit | edit source]- What are other widely used alternatives?
- How does this compare to others?
Scale (Link) | Ease of Use | Languages | Psychometrics | |
---|---|---|---|---|
Text here | Text here | Text here | Text here | Text here |
Text here | Text here | Text here | Text here | Text here |
Text here | Text here | Text here | Text here | Text here |
Text here | Text here | Text here | Text here | Text here |
Development
[edit | edit source]- How were items and content written?
- How was the content checked?
- How were the scales developed?
- Rationally? By experts?
- Statistically?
Administration and scoring
[edit | edit source]General Administration Procedure
[edit | edit source]- Individual or group? Paper, computer, phone?
- Typical time to administer (<--could look at metadata of completion times)
- Age range
- Any special notes for other respondents?
- How to explain purpose to different respondents? (any pro tips for getting valid data?)
- Readability
Computer Administration and Scoring
[edit | edit source]- Notes about any platforms (especially open ones), hardware and software requirements
- REDCap?
- QSF file?
- Links to syntax or other materials
- Pros and cons of computerization
Scoring Instructions
[edit | edit source]- What are the scales? What items go on which scale? Is the total meaningful?
- Could include tables or figures
- What to do with missing responses?
- Are there profiles?
- Percentiles or standard scores?
- Cut scores or Diagnostic Likelihood Ratios?
Interpretation
[edit | edit source]- Scores
- What items go on what scales?
- Raw scores
- Percent of max possible (POMP) scores?
- Percentiles or standard scores?
- Any IRT calibration or special weighting
- If norms
- Sex norms?
- Age norms?
- Informant norms?
- Other demographic considerations
- Evidence of differences across SES, social group, country?
- Integrating information from multiple sources
- Step-by-step sequence for interpretation
Versions
[edit | edit source]Lengths
[edit | edit source]Full length, short forms, alternate forms
Respondents
[edit | edit source]Self (Youth, Adult), Caregiver/Parent, Teacher, Peer/Other, Observer, other
Normative Data
[edit | edit source]Description of the normative sample (if any) or the larger samples that have been published.
Ethnic differences?
Translations
[edit | edit source]Are versions available in additional languages?
Reliability
[edit | edit source]Version | 10-item | 20-item | Full-length |
Mean | 3.08 | 3.05 | 3.04 |
SD | .47 | .47 | .43 |
Range | 2.00 to 4.30 | 2.05 to 4.25 | 2.10 to 3.97 |
Omega Total | .78 | .87 | .89 |
Omega Hierarchical (higher order factor, Schmid-Leiman) | .65 | .73 | .74 |
Omega Group Factors (variance in total score due to 5 factors) | .38 | .22 | .23 |
Observed Mean inter-item correlation | .11 | .15 | .12 |
Observed λ6 | .78 | .86? | .89 |
Projected correlation with full | .72 | .81 | -- |
Observed correlation | .92 | .98 | -- |
Reliability >.8 across range (IRT) | -1.5 to 4.1 | -3.3 to 5.0 | -3.6 to 5.8 |
Discrepancy (Long – Short) | .04ns | .01ns | -- |
SD discrepancy | .xx | .yy | |
95% limits of agreement | .yy to .zz | .yy to .zz | |
Savings in Length (%) | 67% | 33% | 0% |
Standard Error of Measurement | .22 | .17 | .14 |
Standard Error of Difference | .31 | .24 | .20 |
90% Critical Change | .51 | .39 | .33 |
95% Critical Change | .61 | .47 | .40 |
Minimal Important Difference (MID, d ~.5) | .24 | .24 | .22 |
Note. Observed correlations are based on embedded item administration. Standard errors used OmegaTotal as reliability; ns = no significant difference between short and long form.
Validity
[edit | edit source]- Factorial validity
- Convergent and divergent/discriminant validity
- What things should correlate highly? Do they? (= convergent)
- What about measures of other constructs? Do they show lower correlations? (= discriminant)
- Discriminative validity - does it separate groups correctly? (= classification accuracy)
- Other criterion validity
Peer Reviewed Research
[edit | edit source]Special Populations
[edit | edit source]References
[edit | edit source]Links
[edit | edit source]- To PDFs (in OSF?)
- To original site?
- To translations
- To Assessment Center
- To Code
Appendices
[edit | edit source]Code
[edit | edit source]Click here for R code
|
---|
R code goes here |
Click here for SPSS code
|
---|
SPSS code goes here |
Click here for SAS code
|
---|
SAS code goes here |
Codebooks
[edit | edit source]- Marked up PDF
- Google sheet