Dear BIOMCH-L,

I don't understand clearly the TEM(technical error of measurement)

and the reliability.

1. The intra-observer TEM

~~~~~~~~~~~~~~~~~~~~~~

Could you tell me the following procedure that is right or not to find

this TEM and reliability?

TEM = sqrt[(1xD1 + D2xD2 + D3xD3+...... + DNxDN) / 2N]

where D is the difference between measurements, and N is the

number of subjects.

R = 1- [(TEM x TEM) / (SD x SD)]

where SD is the total inter-subject variance.

1. 30 subjects (age 19-20) are measured.

2. Only one body dimension such as stature was measured and repeat

two times on each subject.

3. Find the differece between two repeated measure of them.

4. Then substitute the result 3 into above equation.

5. Find the reliability.

But when I calculate the SD, all the measurement values (two value each

subject) are included. Is it right?

2. The Inter-observer TEM

~~~~~~~~~~~~~~~~~~~~~~

TEM = sqrt[[sum [sum {M(n) x M(n) - sum M(n) x M(n) / K}] /N(K-1)]

Where N is the number of subjects, K is number of determinations

of the variable taken on each subject, M(n) is the nth replicate of

the measurement, where n varies from 1 to K.

1. 5 observer measure 30 subjects, one body dimension is measured

by each observer and no repeat measurement.

Is it K equals to 5 and N equals to 30 and n is 1 to 5?

Thanks in advance.

Best regards,

Laurence

-------------------------------------------------------------------------------

Lau Chung Sing, Laurence

UG student

Department of Industrial Engineering and Engineering Management

Hong Kong University of Science and Technology

Fax: (852)28735456

email: ie_lcsaa@stu.ust.hk

-------------------------------------------------------------------------------

I don't understand clearly the TEM(technical error of measurement)

and the reliability.

1. The intra-observer TEM

~~~~~~~~~~~~~~~~~~~~~~

Could you tell me the following procedure that is right or not to find

this TEM and reliability?

TEM = sqrt[(1xD1 + D2xD2 + D3xD3+...... + DNxDN) / 2N]

where D is the difference between measurements, and N is the

number of subjects.

R = 1- [(TEM x TEM) / (SD x SD)]

where SD is the total inter-subject variance.

1. 30 subjects (age 19-20) are measured.

2. Only one body dimension such as stature was measured and repeat

two times on each subject.

3. Find the differece between two repeated measure of them.

4. Then substitute the result 3 into above equation.

5. Find the reliability.

But when I calculate the SD, all the measurement values (two value each

subject) are included. Is it right?

2. The Inter-observer TEM

~~~~~~~~~~~~~~~~~~~~~~

TEM = sqrt[[sum [sum {M(n) x M(n) - sum M(n) x M(n) / K}] /N(K-1)]

Where N is the number of subjects, K is number of determinations

of the variable taken on each subject, M(n) is the nth replicate of

the measurement, where n varies from 1 to K.

1. 5 observer measure 30 subjects, one body dimension is measured

by each observer and no repeat measurement.

Is it K equals to 5 and N equals to 30 and n is 1 to 5?

Thanks in advance.

Best regards,

Laurence

-------------------------------------------------------------------------------

Lau Chung Sing, Laurence

UG student

Department of Industrial Engineering and Engineering Management

Hong Kong University of Science and Technology

Fax: (852)28735456

email: ie_lcsaa@stu.ust.hk

-------------------------------------------------------------------------------