Question on Intraclass correlation coefficients (ICC): two-way mixed effects

[Ho Hei Cheung]

Hi!
I wish to use ICC to look at test-retest reliability using two sets of data reported by the same individuals. I would need to fill in columns for the dependent variable, target variable, and rater variable according to STATA. Grateful if someone could suggest some codes or help me identify which variable belongs to 'dependent, target and rater variables'. My data is in the following format:

vas1	New_vas1
5	7
6	8
4	5
3	5
4	6

*vas1 - before score; new_vas1 - after score

Thank you so much!

[Joseph Coveney]

Maybe something like the following? (Begin at the "Begin here" comment.)

.ÿ
.ÿversionÿ17.0

.ÿ
.ÿclearÿ*

.ÿ
.ÿinputÿbyte(vas1ÿNew_vas1)

ÿÿÿÿÿÿÿÿÿvas1ÿÿNew_vas1
ÿÿ1.ÿ5ÿÿÿÿÿÿÿ7
ÿÿ2.ÿ6ÿÿÿÿÿÿÿ8
ÿÿ3.ÿ4ÿÿÿÿÿÿÿ5
ÿÿ4.ÿ3ÿÿÿÿÿÿÿ5
ÿÿ5.ÿ4ÿÿÿÿÿÿÿ6
ÿÿ6.ÿend

.ÿ
.ÿ*
.ÿ*ÿBeginÿhere
.ÿ*
.ÿgenerateÿlongÿpidÿ=ÿ_n

.ÿrenameÿvas1ÿvas0

.ÿrenameÿNew_vas1ÿvas1

.ÿ
.ÿquietlyÿreshapeÿlongÿvas,ÿi(pid)ÿj(tim)

.ÿ
.ÿiccÿvasÿpidÿtim,ÿmixed

Intraclassÿcorrelations
Two-wayÿmixed-effectsÿmodel
Consistencyÿofÿagreement

Randomÿeffects:ÿpidÿÿÿÿÿÿÿÿÿÿÿÿÿÿNumberÿofÿtargetsÿ=ÿÿÿÿÿÿÿÿÿ5
ÿFixedÿeffects:ÿtimÿÿÿÿÿÿÿÿÿÿÿÿÿÿNumberÿofÿratersÿÿ=ÿÿÿÿÿÿÿÿÿ2

--------------------------------------------------------------
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿvasÿ|ÿÿÿÿÿÿÿÿICCÿÿÿÿÿÿÿ[95%ÿconf.ÿinterval]
-----------------------+--------------------------------------
ÿÿÿÿÿÿÿÿÿÿÿÿIndividualÿ|ÿÿÿ.9333333ÿÿÿÿÿÿÿ.5024144ÿÿÿÿ.9928452
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿAverageÿ|ÿÿÿ.9655172ÿÿÿÿÿÿÿ.6688093ÿÿÿÿ.9964097
--------------------------------------------------------------
Fÿtestÿthat
ÿÿICC=0.00:ÿF(4.0,ÿ4.0)ÿ=ÿ29.00ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿProbÿ>ÿFÿ=ÿ0.003

Note:ÿICCsÿestimateÿcorrelationsÿbetweenÿindividualÿmeasurements
ÿÿÿÿÿÿandÿbetweenÿaverageÿmeasurementsÿmadeÿonÿtheÿsameÿtarget.

.ÿ
.ÿexit

endÿofÿdo-file


.

[Ho Hei Cheung]

Dear Joseph, thank you so much for your very helpful advice!