Skip to main content
. Author manuscript; available in PMC: 2013 May 31.
Published in final edited form as: Psychiatry Res. 2010 Mar 21;177(0):177–183. doi: 10.1016/j.psychres.2010.02.008

Table 6.

Adjusted correlation values with methods to correct for attenuation.

Timepoint Predictor (X) Dependent (Y) Observed correlation Cronbach's alpha method
 Thorndike's case 2 method
Adjusted correlationa Alpha var 1 Alpha var 2 Adjusted correlation True variancea Observed variance
Baseline HAM-D17 IDS-SR30 0.46 1.21 0.18 0.78 0.80 24.58 9.40
HAM-D17 QIDS-SR16 0.40 1.17 0.18 0.64 0.75 24.58 9.40
CGI-S IDS-SR30 0.27 n.a. 0.66 0.95 0.31
CGI-S QIDS-SR16 0.24 n.a. 0.61 0.95 0.31
Week 10 HAM-D17 IDS-SR30 0.75 0.91 0.76 0.89 0.75 24.58 24.58
HAM-D17 QIDS-SR16 0.73 0.93 0.76 0.80 0.73 24.58 24.58
CGI-S IDS-SR30 0.67 n.a. 0.67 0.95 0.95
CGI-S QIDS-SR16 0.63 n.a. 0.63 0.95 0.95
Month 6 HAM-D17 IDS-SR30 0.70 0.87 0.72 0.88 0.84 24.58 15.54
HAM-D17 QIDS-SR16 0.67 0.91 0.72 0.74 0.82 24.58 15.54
CGI-S IDS-SR30 0.65 n.a. 0.80 0.95 0.60
CGI-S QIDS-SR16 0.64 n.a. 0.79 0.95 0.60

CGI-S, Clinical Global Impressions–Severity; HAM-D17, 17-item Hamilton Rating Scale for Depression; IDS-SR30, 30-item Inventory of Depressive Symptomatology–Self-Report; QIDS-SR16, 16-item Quick Inventory of Depressive Symptomatology–Self-Report.

For the Thorndike's case 2 method, variable 1 serves as the predictor (X), and variable 2 serves as the dependent variable (Y).

a

Note: True variance is an unknown, for this estimate the maximum observed variance (week 10) is used to adjust baseline and month 6 correlations. Week 10 values are reference and unchanged using this approach.