where ? k is the vector of h coefficients corresponding to the k th occupation group. The total number of parameters to be estimated is h(N – 1), since coefficients for each element of x are determined only up to an arbitrary normalization. Thus, the estimated coefficients for each occupation group do not represent marginal probabilities, but indicate the change in the log of odds of being in that occupation instead of in unskilled occupations.
From the positions the brand new coefficients out-of a given variable from the dimensions, we can comprehend the cousin impression of that variable into likelihood of being in particular jobs
The newest quotes of your multinomial logit design signify there was differential entry to job according to gender (look for panamanian women wants to marry american top committee out-of Desk 8). This new coefficient on the female dummy is significantly distinctive from zero in five industry groups together with sign may differ across the occupations. Ceteris paribus, the likelihood of being an employer otherwise plant and you can server agent is leaner (the latest coefficient to the women dummy was bad and tall) additionally the likelihood of getting an expert, professional and you may user top-notch otherwise a good clerical staff member is actually high (the coefficient into the feminine dummy is actually positive and you can tall) for ladies compared to the dudes.
There isn’t any factor ranging from people in the likelihood of are interested because the solution and you may conversion gurus or becoming into the hobby and you may related change
To measure the impact of differences in “tastes”, hiring prejudices, or other unobservable factors on the occupational distribution, we estimate a separate model of occupational attainment for men (see lower panel of Table 8). Employing these estimates, we obtain the predicted distribution for each sex as follows: substitute the sample data for each sex into the estimated model, producing for each individual a vector of predicted probabilities of belonging to each of the eight occupation groups, and calculate the mean of the predicted probabilities for each occupation after summing over observations. For men this estimation yields a predicted distribution which is identical to their actual sample distribution, i.e., M p = M a . The difference in the predicted distributions of men and women (M a –W p ), is the ‘explained’ component due to difference in characteristics, and the residual difference, (W p –W a ), is the ‘unexplained’ component due to differences in tastes, hiring prejudices, or other unobservable factors.
The outcome of decomposition get it done are shown within the upper committee away from Desk 9. I interest primarily into decompositions for the highest variations in the real decide to try withdrawals. Brand new get it done signifies that when the you will find zero differential the means to access field by gender, the fresh new ratio of women interested due to the fact masters, mechanics and affiliate masters, and you can clerks might be much smaller than noticed, while the proportion of females interested due to the fact managers and you can bush providers will be considerably higher than seen (column 5). The rest of the “unexplained” role makes up about to 73-78 per cent of your seen difference between men and women in new size engaged in elite and you may clerical jobs, and you will 97 % in the example of plant and you can servers workers (line seven). Considering the character of these employment, this almost certainly reflects “supply-side” differences in choice of women to possess light-collar jobs. For professionals and older officials, the remaining “unexplained” differences (0.0269) is simply higher than this new observed improvement (0.0198). When the there was zero differential access to field of the gender, new ratio from inside the managerial positions might possibly be large for women than for men. It is hard to help you ascribe that it looking so you’re able to “supply-side” difference in choices.