Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# RE: st: Intraclass correlation coefficient

 From "Sofia Ramiro" To Subject RE: st: Intraclass correlation coefficient Date Fri, 14 Oct 2011 00:33:45 +0200

```Thanks for the answers.
Not totally sure i understood, but from what I read it seemed to me that I
should do loneway var1 var2
I did it and I got totally different results from the ones obtained in SPSS
- ICC 0.805 (stata) and only one ICC, and 0.811 single measures ICC and
0.896 average measures ICC in the case of SPSS. It's exactly the same
dataset and same variables. I don't understand
Do I need to add any other specifications in stata? And getting only one ICC
which is this one - single measures or average measures?
Thanks
S.

. loneway newprogsr1 newprogcs1

One-way Analysis of Variance for newprogsr1:

Number of obs =       187
R-squared =    0.7747

Source                SS         df      MS            F     Prob > F
-------------------------------------------------------------------------
Between newprogcs1      2067.369     15     137.8246     39.20     0.0000
Within newprogcs1      601.26199    171     3.516152
-------------------------------------------------------------------------
Total                   2668.631    186    14.347479

Intraclass       Asy.
correlation      S.E.       [95% Conf. Interval]
------------------------------------------------
0.80585     0.10581       0.59847     1.01323

Estimated SD of newprogcs1 effect       3.820238
Estimated SD within newprogcs1          1.875141
Est. reliability of a newprogcs1 mean    0.97449
(evaluated at n=9.20)

Intraclass Correlation Coefficient
95% Confidence Interval			F Test with True
Value 0
Intraclass Correlationa	Lower Bound	Upper Bound	Value	df1
df2	Sig
Single Measures	,811b	,755	,855	9,576	186	186	,000
Average Measures	,896c	,861	,922	9,576	186	186
,000
Two-way mixed effects model where people effects are random and measures
effects are fixed.
a. Type C intraclass correlation coefficients using a consistency
definition-the between-measure variance is excluded from the denominator
variance.
b. The estimator is the same, whether the interaction effect is present or
not.
c. This estimate is computed assuming the interaction effect is absent,
because it is not estimable otherwise.

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Cameron McIntosh
Sent: quinta-feira, 13 de Outubro de 2011 19:06
To: STATA LIST
Subject: RE: st: Intraclass correlation coefficient

Following up on Ronán's comment, I might suggest having a look at:

Rousson, V. (2011). Assessing inter-rater reliability when the raters are
fixed: Two concepts and two estimates. Biometrical Journal, 53(3), 477?490.

Rousson, V., Gasser, T., & Seifert, B. (2002). Assessing intrarater,
interrater and test?retest reliability of continuous measurements.
Statistics in Medicine, 21(22), 3431?3446.

Gamer, M., Lemon, J., Fellows, I., & Singh, P. (September 20, 2011).
Various Coefficients of Interrater Reliability and Agreement: Package
?irr?. Version 0.83.

http://cran.r-project.org/web/packages/irr/irr.pdf

http://cran.r-project.org/web/packages/irr/index.html

Lu, L., & Shara, N. (2007). Reliability analysis: Calculate and Compare
Intra-class Correlation Coefficients (ICC) in SAS. NESUG 2007:
Statistics and Data Analysis.

http://www.nesug.org/proceedings/nesug07/sa/sa13.pdf

Cohen, A., & Doveh, E. (2005). More on the Comparison of Intra-Class
Correlation Coefficients (ICCs) as Measures of Homogeneity. In F.
Dansereau & F.J. Yammarino (Eds.), Multi-Level Issues in Strategy and
Methods (Research in Multi-Level Issues, vol. 4, pp.455-459).
Emerald Group Publishing Limited.

Cam

> From: rconroy@rcsi.ie
> To: statalist@hsphsun2.harvard.edu
> Date: Thu, 13 Oct 2011 17:40:13 +0100
> Subject: Re: st: Intraclass correlation coefficient
>
> On 2011 DFómh 13, at 14:10, Sofia Ramiro wrote:
>
> > Hello,
> >
> > Does anyone know how to calculate the ICC between 2 variables?
> > I am analyzing the inter-reader reliability, and only know how to
calculate
> > ICCs with SPSS. However, does not make much sense to use Stata and
switch
> > back to SPSS just because of this small part of the analysis
> > I appreciate your help
>
> There are three 'flavours' of intraclass correlation corresponding to
different hypotheses. Type one intraclass correlations are handled by
-loneway-. The others are handled by -icc23-, written by Paul F. Visintainer
and Luis C.Orozco.
>
> You need to know which flavour you require.
>
> Ronán Conroy
> rconroy@rcsi.ie
> Associate Professor
> Division of Population Health Sciences
> Royal College of Surgeons in Ireland
> Beaux Lane House
> Dublin 2
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```