Is "linear-by-linear association" in SPSS one more name because that the chi-squared test because that trend? If not, what is it?

You are watching: Linear by linear association

It is. You could have a look at IBM"s support web page for SPSS,where it is stated in a technote ~ above the Chi² test:"The Crosstabs procedure contains the Mantel-Haenszel check of trend among its chi-square test statistics. ... The MH test for trend will certainly be printed in the "Chi-Square Tests" table and also labelled "Linear-by-Linear Association"."

see: https://www-304.ibm.com/support/docview.wss?uid=swg21477269

See more: Data Is Information That Has Been Shaped Into A Form That Is Meaningful To Human Beings.

As a previous reply mentioned, yes it is and the technical summary is in ~ SPSS"s support page: https://www-304.ibm.com/support/docview.wss?uid=swg21477269

This is a valuable statistic for those who recognize it. Intend we investigate whether 78 employees" promotion (yes/no) is pertained to their performance ranking in the previous year (1-4, 1=low), as follows:

Ranking 1: Not supported 17, promoted 2, complete 19. Ranking 2: Not advocated 16, advocated 4, full 20.Ranking 3: Not advocated 14, advocated 6, total 20.Ranking 4: Not promoted 10, promoted 9, full 19.

SPSS reflects a significant linear-by-linear combination (p=.008) showing that there is a far-ranging association in between the ranking and also being promoted.

Some valuable details of how this functions are:1. The check relates to the odds. Odds are provided for their statistics properties, and are not quite the exact same as probabilities. For ranking 1, the odds the being promoted are 2:17, together opposed come the probability which is 2:19.2. Then, the check is on the odds ratios; e.g. If you move from rank 1 to rank 2, the odds ratio is 4:16/2:17 = 0.250/0.118 = 2.12. (The null hypothesis is the the odds proportion is 1, i.e. A readjust in ranks makes no distinction to the odds.)3. The procedure presumes that the odds ratios (in the population) are the very same for all procedures (i.e. If relocating from location 1 to location 2 doubles the odds the promotion, relocating from rank 2 to rank 3 would certainly also double the odds of promotion). The is why there is just 1 level of freedom. (This assumption is recognized as "linearity in the logit".)4. The check is because of this conceptually the same (and offers a similar answer) come doing logistic regression with just one covariate. (In logistic regression, "covariate" means a variable choose this one). In this situation the covariate would certainly be ranking, and the DV would be promotion decision.