The CSV score output file provides the scores for each examinee, separated by item type. The columns in this file are as follows:
- Sequence – The row number of the examinee in the data file.
- ID – Examinee ID from the data file.
- Scored items – The total number-correct (raw) score for all scored items across all domains.
- All items – The total number-correct score for all scored items across all domains plus all of the pretest items included in the test.
- Pretest items – Provides the number-correct scores for the pretest items only.
- Scored Proportion Correct – The proportion of scored items correct across all domains. This value is only output if there are no scored polytomous items.
- Rank – The rank of the examinee’s score, from 1 to N, which is calculated as the number of examinees with total scores for the scored items (column 1) greater than or equal to the given examinee. Examinees that are tied with the same score receive the lowest rank available (e.g., Examinees 4 and 8 have ranks of 4,167). Examinee(s) with the highest score will receive a rank of 1.
- Percentile – The percentage of examinees whose total score falls below the given score.
- Group – The score group an examinee is classified into is based on total score for all the scored items. The number of score groups is determined by the value set in the “Create X score groups for the quantile plots / summary table” box on the Output Options tab.
- Domain X – The final column(s) in the scores output file contain the domain scores separately for each domain. If more than one domain was specified in the Item Control File, then the total number-correct scores calculated separately for each domain would be found after the “Group” column. If the test has only one domain, this will not be included, as it will be the same as the “Scored items” column.
- Scaled Total Score – If scaled scores were computed for total score these will appear.
- Scaled Domain Score X – If scaled scores were requested for each domain – and there is more than one domain – these will be provided.
- Classification – If dichotomous classification was performed the results of the classification will be provided. Classification can be performed with the raw total number-correct scores or the scaled total number correct score.
- CSEM III – The conditional standard error of measurement Formula III from Lord (1984) if there are no scored polytomous items.
- CSEM IV – The conditional standard error of measurement formula IV from Lord (1984) if there are no scored polytomous items. Across all observed scores, this formula is most comparable to the classical SEM found in Table 4.2.
The following image is an example of 10 examinees, with no scaled scoring.
Article is closed for comments.