| Tag | Value |
|---|---|
| file | Reliability_eur-reliability-108-en_eur-reliability-108-en |
| name | eur-reliability-108-en |
| section | Reliability/Analysis/Cronbach's alpha |
| type | string |
| solution | "" |
| Type | Conceptual |
| Program | |
| Language | English |
| Level | Statistical Literacy |
How is it possible that although alpha places a lower limit on the size of the estimated reliability, in practice it is possible to get an alpha that is higher than the true reliability?
It is possible that although alpha places a lower limit on the size of the estimated reliability, in practice one gets an alpha that is higher than the true reliability. This is because the alpha is calculated using a sample, and not the complete population. So the sample alpha may be large (even larger than the true reliability ) due to sampling variability.