2007
DOI: 10.1016/j.sapharm.2006.04.001
|View full text |Cite
|
Sign up to set email alerts
|

Missing data on the Center for Epidemiologic Studies Depression Scale: A comparison of 4 imputation techniques

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
100
0
1

Year Published

2007
2007
2022
2022

Publication Types

Select...
10

Relationship

0
10

Authors

Journals

citations
Cited by 132 publications
(102 citation statements)
references
References 42 publications
1
100
0
1
Order By: Relevance
“…We first computed individual NP-level mean scores on the AIP and NP-AR scales for respondents who completed more than 70 % of the scale items, 39 then aggregated the scores of all NPs from each practice and computed the practice-level AIP and NP-AR scores and calculated intra-class correlations (ICCs). We created an NP-level TW scale by calculating the scale mean.…”
Section: Resultsmentioning
confidence: 99%
“…We first computed individual NP-level mean scores on the AIP and NP-AR scales for respondents who completed more than 70 % of the scale items, 39 then aggregated the scores of all NPs from each practice and computed the practice-level AIP and NP-AR scores and calculated intra-class correlations (ICCs). We created an NP-level TW scale by calculating the scale mean.…”
Section: Resultsmentioning
confidence: 99%
“…Also, the Zung depression rating scale as a whole was not predictive for the development of depression in substance use patients, at least after exclusion of the patients that were already depressed at baseline. In other studies the baseline value was found to be predictive for patients with the Minnesota Multiphasic Personality Inventory [22] , the Hamilton Depression Rating Scale score (HDRSS) [23] and the Centre for Epidemiological StudiesDepression Scale, CES-D [24] higher than a certain cut-off level [21,25,26] . However, in these studies people that were depressed at baseline were not always excluded [25,26] .…”
Section: Discussionmentioning
confidence: 99%
“…Missing data were considered missing at random, and comprised only 1.2 % and 3.2 % of the parent and teacher data sets. Therefore, Pearson mean imputation was used to replace missing scale items with mean values, providing at least 80 % of the data were available (Bono et al 2007). Next, baseline sample characteristics were assessed for comparability between intervention and control participants.…”
Section: Analytic Strategymentioning
confidence: 99%