Hỏi đáp

# Cronbach’S Alpha Là Gì, Cronbachs Alpha In Spss Statistics

Cronbach’s alpha is a measure of internal consistency, that is, how closelyrelated a set of items are as a group. It is considered to be a measureof scale reliability. A “high” value for alphadoes not imply that the measure is unidimensional. If, in addition to measuringinternal consistency, you wish to provide evidence that the scale in question isunidimensional, additional analyses can be performed. Exploratory factoranalysis is one method of checking dimensionality. Technically speaking, Cronbach’s alpha is nota statistical test – it is a coefficient of reliability (or consistency).

Chúng ta đang xem: Cronbach’s alpha là gì

Cronbach’s alpha can be writtenas a function of the number of test items and the average inter-correlationamong the items. Below, for conceptual purposes, we show the formula forthe Cronbach’s alpha:

\$\$ alpha = frac{N bar{c}}{bar{v} + (N-1) bar{c}}\$\$

Here \$N\$ is equal to the number of items, \$bar{c}\$ is the average inter-item covariance among the items and\$bar{v}\$ equals the average variance.

One can see from this formula that if you increase the number of items, you increase Cronbach’s alpha.Additionally, if the average inter-item correlation is low, alpha will be low. As the average inter-item correlation increases, Cronbach’s alpha increases as well (holding the number of items constant).

## An example

Let’s work through an example of how to compute Cronbach’s alpha using SPSS, and how to kiểm tra the dimensionalityof the scale using factor analysis. For this example, we will use a dataset that contains four test items – q1, q2, q3 and q4. You can tải về the dataset by clicking on https://a.viettingame.vn/wp-content/uploads/2016/02/alpha.sav.To compute Cronbach’s alpha for all four items – q1, q2, q3, q4 – use the reliability command:

RELIABILITY /VARIABLES=q1 q2 q3 q4.Here is the resulting output from the above syntax: The alpha coefficient for the four items is .839, suggesting that the itemshave relatively high internal consistency. (Notethat a reliability coefficient of .70 or higher is considered“acceptable” in most social science research situations.)

### Hand calculation of Cronbach’s Alpha

For demonstration purposes, here is how to calculate the results above by hand. In SPSS, you can obtain covariances by going to AnalyzeCorrelateBivariate. Then shift q1, q2,q3 and q4 to the Variables box and nhấp chuột Options. Under Statistics, kiểm tra Cross-product deviations and covariances. Nhấp chuột Continue and OK to obtain output. Below you will see a condensed version of the output. Notice that the diagonals (in bold) are the variances and the off-diagonals are the covariances. We only need to consider the covariances on the lower left triangle because this is a symmetric matrix.

 q1 q2 q3 q4 q1 Covariance 1.168 .557 .574 .673 q2 Covariance .557 1.012 .690 .720 q3 Covariance .574 .690 1.169 .724 q4 Covariance .673 .720 .724 1.291

Recall that \$N=4\$ is equal to the number of items, \$bar{c}\$ is the average inter-item covariance among the items and\$bar{v}\$ equals the average variance. Using the information from the table above, we can calculate each of these components via the following:

\$\$bar{v} = (1.168 + 1.012 + 1.169 + 1.291)/4 = 4.64 / 4 = 1.16.\$\$

\$\$bar{c} = (0.557 + 0.574 + 0.690 + 0.673 + 0.720 + 0.724)/6 = 3.938 / 6 = 0.656.\$\$

\$\$ alpha = frac{4 (0.656)}{(1.16) + (4-1) (0.656)}=2.624/3.128=0.839.\$\$

The results match our SPSS obtained Cronbach’s Alpha of 0.839.

In addition to computing the alpha coefficient of reliability, we might alsowant to investigate the dimensionality of the scale. We can use the factorcommand to do this:

FACTOR /VARIABLES q1 q2 q3 q4 /FORMAT SORT BLANK(.35).Here is the resulting outputfrom the above syntax: Looking at the table labeled Total Variance Explained, we see that the eigen value forthe first factor is quite a bit larger than the eigen value for the next factor (2.7versus 0.54).Additionally, the first factor accounts for 67% of the total variance. This suggests that thescale items are unidimensional. 