Dear community members,
With reference to thread https://www.statalist.org/forums/for...-of-i-category .
I have estimated using xthybrid
Code:
. xthybrid Positive_disc01 stud_SCST stud_OBC Teach_SCST Teach_OBC Teach_nature_1 Teach_nature_2 Teach_gender_1 course1_ > com course1_eco course1_eng course1_hin course1_his course1_mat course1_pol sem_1 sem_2 sem_3 attendence_percent , clusterid > (group_teacher_paper) se test p star The variable 'Teach_SCST' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'Teach_SCST' is within clusters] The variable 'Teach_OBC' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'Teach_OBC' is within clusters] The variable 'Teach_nature_1' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'Teach_nature_1' is within clusters] The variable 'Teach_nature_2' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'Teach_nature_2' is within clusters] The variable 'Teach_gender_1' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'Teach_gender_1' is within clusters] The variable 'course1_com' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'course1_com' is within clusters] The variable 'sem_1' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'sem_1' is within clusters] The variable 'sem_2' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'sem_2' is within clusters] The variable 'sem_3' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'sem_3' is within clusters] Hybrid model. Family: gaussian. Link: identity. +--------------------------------------+ | Variable | model | |----------------------+---------------| | Positive_disc01 | | | R__Teach_SCST | -0.0356 | | R__Teach_OBC | 0.0240 | | R__Teach_nature_1 | 0.0118 | | R__Teach_nature_2 | (omitted) | | R__Teach_gender_1 | -0.0540 | | R__course1_com | -0.3539** | | R__sem_1 | -0.1937** | | R__sem_2 | -0.1524* | | R__sem_3 | -0.0967* | | W__stud_SCST | -0.0297** | | W__stud_OBC | -0.0224* | | W__course1_eco | 0.0318 | | W__course1_eng | 0.0765 | | W__course1_hin | 0.0723 | | W__course1_his | -0.0204 | | W__course1_mat | 0.0190 | | W__course1_pol | (omitted) | | W__attendence_perc~t | 0.0016*** | | B__stud_SCST | 0.0838 | | B__stud_OBC | -0.4088 | | B__course1_eco | -0.2559* | | B__course1_eng | -0.5010*** | | B__course1_hin | -0.1966*** | | B__course1_his | -0.3835*** | | B__course1_mat | -0.2841* | | B__course1_pol | (omitted) | | B__attendence_perc~t | 0.0068*** | | _cons | 0.5530** | |----------------------+---------------| | var(_cons[group~r])| | | _cons | 0.0442*** | |----------------------+---------------| | var(e.Positive_di~01)| | | _cons | 0.1664*** | |----------------------+---------------| | Statistics | | | ll | -5530.5953 | | chi2 | 208.1611 | | p | 0.0000 | | aic | 11115.1906 | | bic | 11310.1144 | +--------------------------------------+ Legend: * p<.05; ** p<.01; *** p<.001 Level 1: 10091 units. Level 2: 201 units. Tests of the random effects assumption: _b[B__stud_SCST] = _b[W__stud_SCST]; p-value: 0.7050 _b[B__stud_OBC] = _b[W__stud_OBC]; p-value: 0.4125 _b[B__course1_eco] = _b[W__course1_eco]; p-value: 0.0117 _b[B__course1_eng] = _b[W__course1_eng]; p-value: 0.0000 _b[B__course1_hin] = _b[W__course1_hin]; p-value: 0.0010 _b[B__course1_his] = _b[W__course1_his]; p-value: 0.0053 _b[B__course1_mat] = _b[W__course1_mat]; p-value: 0.0215 _b[B__course1_pol] = _b[W__course1_pol]; p-value: . _b[B__attendence_percent] = _b[W__attendence_percent]; p-value: 0.0093
In my research, I am mainly interested to know if students who have a disadvantaged social identity (i.e., stud_SCST), do they have a lower chance of getting 1 in Positive_disc01?
Concern-1 : What conclusion am I able to draw from the above regression?
Concern-2 : Am I required to perform some other type of regression.?
Concern-3 :My concern is also with the _b[B__stud_SCST] = _b[W__stud_SCST]; p-value: 0.7050. I am not sure what it implies. Does it mean I need to try some other model type...!!?
After the first regression I ran, analogous correlated random-effects model
Code:
. xthybrid Positive_disc01 stud_SCST stud_OBC Teach_SCST Teach_OBC Teach_nature_1 Teach_nature_2 Teach_gender_1 course1_ > com course1_eco course1_eng course1_hin course1_his course1_mat course1_pol sem_1 sem_2 sem_3 attendence_percent , clusterid > (group_teacher_paper) se test cre p star The variable 'Teach_SCST' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'Teach_SCST' is within clusters] The variable 'Teach_OBC' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'Teach_OBC' is within clusters] The variable 'Teach_nature_1' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'Teach_nature_1' is within clusters] The variable 'Teach_nature_2' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'Teach_nature_2' is within clusters] The variable 'Teach_gender_1' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'Teach_gender_1' is within clusters] The variable 'course1_com' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'course1_com' is within clusters] The variable 'sem_1' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'sem_1' is within clusters] The variable 'sem_2' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'sem_2' is within clusters] The variable 'sem_3' does not vary sufficiently within clusters and will not be used to create additional regressors. [~0% of the total variance in 'sem_3' is within clusters] Correlated random effects model. Family: gaussian. Link: identity. +--------------------------------------+ | Variable | model | |----------------------+---------------| | Positive_disc01 | | | R__Teach_SCST | -0.0356 | | R__Teach_OBC | 0.0240 | | R__Teach_nature_1 | 0.0118 | | R__Teach_nature_2 | (omitted) | | R__Teach_gender_1 | -0.0540 | | R__course1_com | -0.3539** | | R__sem_1 | -0.1937** | | R__sem_2 | -0.1524* | | R__sem_3 | -0.0967* | | W__stud_SCST | -0.0297** | | W__stud_OBC | -0.0224* | | W__course1_eco | 0.0318 | | W__course1_eng | 0.0765 | | W__course1_hin | 0.0723 | | W__course1_his | -0.0204 | | W__course1_mat | 0.0190 | | W__course1_pol | (omitted) | | W__attendence_perc~t | 0.0016*** | | D__stud_SCST | 0.1135 | | D__stud_OBC | -0.3863 | | D__course1_eco | -0.2877* | | D__course1_eng | -0.5775*** | | D__course1_hin | -0.2689*** | | D__course1_his | -0.3632** | | D__course1_mat | -0.3031* | | D__course1_pol | (omitted) | | D__attendence_perc~t | 0.0052** | | _cons | 0.5530** | |----------------------+---------------| | var(_cons[group~r])| | | _cons | 0.0442*** | |----------------------+---------------| | var(e.Positive_di~01)| | | _cons | 0.1664*** | |----------------------+---------------| | Statistics | | | ll | -5530.5953 | | chi2 | 208.1611 | | p | 0.0000 | | aic | 11115.1906 | | bic | 11310.1144 | +--------------------------------------+ Legend: * p<.05; ** p<.01; *** p<.001 Level 1: 10091 units. Level 2: 201 units. Tests of the random effects assumption: _b[D__stud_SCST] = 0; p-value: 0.7050 _b[D__stud_OBC] = 0; p-value: 0.4125 _b[D__course1_eco] = 0; p-value: 0.0117 _b[D__course1_eng] = 0; p-value: 0.0000 _b[D__course1_hin] = 0; p-value: 0.0010 _b[D__course1_his] = 0; p-value: 0.0053 _b[D__course1_mat] = 0; p-value: 0.0215 _b[D__course1_pol] = 0; p-value: . _b[D__attendence_percent] = 0; p-value: 0.0093
