Navigating the Unfair Running Race: A Quantitative Analysis of Career Success Factors in Software Engineering Using Open Datasets
Abstract
Context. Gender diversity in software development teams impacts software quality and team climate; however, women are still underrepresented. In the competitive landscape of software engineering careers, gender disparities often resemble an unequal running race in which women face systemic obstacles that hinder their progression and recognition. To address this imbalance, we need to attract more women to the field and ensure they remain in these roles. One key aspect of retaining women in software development teams is their perception of career success, which can be influenced by numerous factors.Objective. This study aims to understand factors that influence career success in Software engineering, with a particular focus on the subjective perception of career success and how gender impacts it. Research Method. We employed a quantitative approach using open and publicly available datasets produced by other researchers. We began with a literature review to develop a theoretical model and formulate hypotheses. Afterward, we selected appropriate datasets and tested our hypotheses. Results. We developed a model that includes career progress satisfaction (the dependent variable) and three categories of independent variables: human capital (experience, education), socio-demographic status (gender, age), and organizational environment (no organizational difficulties). We confirmed all proposed hypotheses, demonstrating that gender directly affects our dependent variable and moderates all proposed relationships.Conclusions. This research presents a model that companies and individuals can utilize to define strategies for career progress. Additionally, the model can assist researchers in determining the focus of their investigations. We also provide an example of effectively reusing a publicly available dataset.
Full text
VIF - Career Progress - > SocioDemographic Variables Entered/Removeda Model Variables Entered Variables Removed Method 1 Age, Genderb.Enter Dependent Variable: Career Progressa. All requested variables entered.b. Coefficientsa Model Collinearity Statistics Tolerance VIF 1Gender Age .973 1.028 .973 1.028 Dependent Variable: Career Progress a. Nominal Regression - Career Progress - > SocioDemographic Warnings There are 2 (8.3%) cells (i.e., dependent variable levels by subpopulations) with zero frequencies. Measures of Monotone Association table is not generated because the dependent variable does not have exactly two levels. Case Processing Summary NMarginal Percentage Career Progress Below the expectations Neutral According to the expectations Above the expectations Gender Men Women Age Less than 20 years old Between 21 to 30 years old More than 30 years old Valid Missing Total Subpopulation 75 22.7% 31 9.4% 156 47.1% 69 20.8% 176 53.2% 155 46.8% 15 4.5% 167 50.5% 149 45.0% 331 100.0% 0 331 6 Page 1
Model Fitting Information Model Model Fitting Criteria Likelihood Ratio Tests AIC BIC -2 Log Likelihood Chi-Square df Sig. Intercept Only Final 137.907 149.313 131.907 94.662 140.287 70.662 61.245 9<.001 Pseudo R-Square Cox and Snell Nagelkerke McFadden .169 .184 .075 Likelihood Ratio Tests Effect Model Fitting Criteria Likelihood Ratio Tests AIC of Reduced Model BIC of Reduced Model -2 Log Likelihood of Reduced Model Chi-Square df Sig. Intercept Gender Age 94.662 140.287 70.662a.000 0 . 143.118 177.337 125.118 54.456 3<.001 85.628 108.441 73.628 2.967 6.813 The chi-square statistic is the difference in -2 log-likelihoods between the final model and a reduced model. The reduced model is formed by omitting an effect from the final model. The null hypothesis is that all parameters of that effect are 0. This reduced model is equivalent to the final model because omitting the effect does not increase the degrees of freedom. a. Page 2
Parameter Estimates Career Progressa BStd. Error Wald df Sig. Neutral Intercept [Gender=1] [Gender=2] [Age=1] [Age=2] [Age=3] According to the expectations Intercept [Gender=1] [Gender=2] [Age=1] [Age=2] [Age=3] Above the expectations Intercept [Gender=1] [Gender=2] [Age=1] [Age=2] [Age=3] -1.689 .455 13.776 1<.001 1.809 .479 14.285 1<.001 6.102 0b. . 0 . . .228 1.290 .031 1.860 1.256 -.184 .455 .165 1.685 .832 0b. . 0 . . .386 .260 2.194 1.139 .850 .308 7.621 1.006 2.339 0b. . 0 . . .602 .840 .513 1.474 1.826 .014 .298 .002 1.963 1.014 0b. . 0 . . -1.598 .416 14.765 1<.001 2.692 .434 38.459 1<.001 14.766 0b. . 0 . . 1.099 .943 1.358 1.244 3.001 -.260 .378 .471 1.492 .771 0b. . 0 . . Parameter Estimates Career Progressa Exp(B) 95% Confidence Interval for Exp (B) Lower Bound Upper Bound Neutral Intercept [Gender=1] [Gender=2] [Age=1] [Age=2] [Age=3] According to the expectations Intercept [Gender=1] [Gender=2] [Age=1] [Age=2] [Age=3] Above the expectations Intercept [Gender=1] [Gender=2] [Age=1] [Age=2] [Age=3] 6.102 2.389 15.590 ... 1.256 .100 15.747 .832 .341 2.027 ... 2.339 1.279 4.277 ... 1.826 .352 9.480 1.014 .566 1.818 ... 14.766 6.306 34.580 ... 3.001 .473 19.052 .771 .368 1.618 ... Page 3
The reference category is: Below the expectations.a. This parameter is set to zero because it is redundant.b. Classification Observed Predicted Below the expectations Neutral According to the expectations Above the expectations Below the expectations Neutral According to the expectations Above the expectations Overall Percentage 0 0 73 20.0% 0 0 31 00.0% 0 0 153 398.1% 0 0 66 34.3% 0.0% 0.0% 97.6% 2.4% 47.1% Classification Observed Predicted Percent Correct Below the expectations Neutral According to the expectations Above the expectations Overall Percentage 0.0% 0.0% 98.1% 4.3% 47.1% Page 4
Observed and Predicted Frequencies Age Gender Career Progress Frequency Observed Predicted Less than 20 years old Men Below the expectations Neutral According to the expectations Above the expectations Women Below the expectations Neutral According to the expectations Above the expectations Between 21 to 30 years old Men Below the expectations Neutral According to the expectations Above the expectations Women Below the expectations Neutral According to the expectations Above the expectations More than 30 years old Men Below the expectations Neutral According to the expectations Above the expectations Women Below the expectations Neutral According to the expectations Above the expectations 2.453 2.367 0.641 -.835 32.845 .114 34.061 -.750 01.547 -1.409 1.359 1.099 44.155 -.119 2.939 1.176 12 9.443 .892 98.848 .055 29 32.947 -.928 23 21.762 .317 31 33.557 -.551 55.152 -.069 54 50.053 .816 45.238 -.556 711.104 -1.311 13 12.511 .148 42 38.207 .794 33 33.177 -.038 23 18.896 1.171 33.489 -.271 24 27.793 -1.033 43.823 .094 Page 5
Observed and Predicted Frequencies Age Gender Career Progress Frequency Percentage Pearson Residual Observed Less than 20 years old Men Below the expectations Neutral According to the expectations Above the expectations Women Below the expectations Neutral According to the expectations Above the expectations Between 21 to 30 years old Men Below the expectations Neutral According to the expectations Above the expectations Women Below the expectations Neutral According to the expectations Above the expectations More than 30 years old Men Below the expectations Neutral According to the expectations Above the expectations Women Below the expectations Neutral According to the expectations Above the expectations 2.367 25.0% 5.7% -.835 0.0% 8.0% .114 37.5% 35.6% -.750 37.5% 50.8% -1.409 0.0% 22.1% 1.099 14.3% 5.1% -.119 57.1% 59.4% 1.176 28.6% 13.4% .892 16.4% 12.9% .055 12.3% 12.1% -.928 39.7% 45.1% .317 31.5% 29.8% -.551 33.0% 35.7% -.069 5.3% 5.5% .816 57.4% 53.2% -.556 4.3% 5.6% -1.311 7.4% 11.7% .148 13.7% 13.2% .794 44.2% 40.2% -.038 34.7% 34.9% 1.171 42.6% 35.0% -.271 5.6% 6.5% -1.033 44.4% 51.5% .094 7.4% 7.1% Page 6
Observed and Predicted Frequencies Age Gender Career Progress Percentage Predicted Less than 20 years old Men Below the expectations Neutral According to the expectations Above the expectations Women Below the expectations Neutral According to the expectations Above the expectations Between 21 to 30 years old Men Below the expectations Neutral According to the expectations Above the expectations Women Below the expectations Neutral According to the expectations Above the expectations More than 30 years old Men Below the expectations Neutral According to the expectations Above the expectations Women Below the expectations Neutral According to the expectations Above the expectations 5.7% 8.0% 35.6% 50.8% 22.1% 5.1% 59.4% 13.4% 12.9% 12.1% 45.1% 29.8% 35.7% 5.5% 53.2% 5.6% 11.7% 13.2% 40.2% 34.9% 35.0% 6.5% 51.5% 7.1% The percentages are based on total observed frequencies in each subpopulation. VIF - Career Progress - >Human Capital Variables Entered/Removeda Model Variables Entered Variables Removed Method 1 Exp, Ed. Levelb.Enter Dependent Variable: Career Progressa. All requested variables entered.b. Page 7
Coefficientsa Model Collinearity Statistics Tolerance VIF 1Ed. Level Exp .775 1.290 .775 1.290 Dependent Variable: Career Progress a. Nominal Regression - Career Progress - > Human Capital Warnings There are 33 (34.4%) cells (i.e., dependent variable levels by subpopulations) with zero frequencies. Unexpected singularities in the Hessian matrix are encountered. This indicates that either some predictor variables should be excluded or some categories should be merged. The NOMREG procedure continues despite the above warning(s). Subsequent results shown are based on the last iteration. Validity of the model fit is uncertain. Measures of Monotone Association table is not generated because the dependent variable does not have exactly two levels. Page 8
Case Processing Summary NMarginal Percentage Career Progress Below the expectations Neutral According to the expectations Above the expectations Ed. Level High School Bachelor Student Bachelor Degree Master Student Master Degree PhD Student PhD Degree Exp Less than 1 year Between 1 to 10 years Between 10 to 15 years More than 15 years Valid Missing Total Subpopulation 75 22.7% 31 9.4% 156 47.1% 69 20.8% 51.5% 90 27.2% 158 47.7% 29 8.8% 30 9.1% 11 3.3% 82.4% 34 10.3% 175 52.9% 51 15.4% 71 21.5% 331 100.0% 0 331 24a The dependent variable has only one value observed in 6 (25.0%) subpopulations. a. Model Fitting Information Model Model Fitting Criteria Likelihood Ratio Tests AIC BIC -2 Log Likelihood Chi-Square df Sig. Intercept Only Final 193.321 204.727 187.321 204.375 318.439 144.375 42.946 27 .026 Pseudo R-Square Cox and Snell Nagelkerke McFadden .122 .133 .052 Page 9
Observed and Predicted Frequencies Exp Ed. Level Career Progress Frequency Observed Predicted Between 1 to 10 years Master Degree Below the expectations Neutral According to the expectations Above the expectations PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations Between 10 to 15 years Bachelor Student Below the expectations Neutral According to the expectations Above the expectations Bachelor Degree Below the expectations Neutral According to the expectations Above the expectations Master Student Below the expectations Neutral According to the expectations Above the expectations Master Degree Below the expectations Neutral According to the expectations Above the expectations PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations 2.918 1.287 0.535 -.786 21.810 .190 0.736 -.950 0.378 -.658 2.739 1.690 0.913 -1.146 1.970 .037 0.323 -.691 0.290 -.639 1.137 2.513 0.250 -.578 0.087 -.310 0.086 -.307 1.522 .957 0.304 -.661 76.185 .364 33.014 -.008 14 14.390 -.137 99.411 -.158 0.857 -1.095 1.269 1.477 1.925 .093 1.948 .064 11.300 -.284 11.347 -.324 33.332 -.229 43.021 .691 0.199 -.462 0.692 -.949 1.626 .532 21.482 .598 1.371 1.143 1.591 .634 0.204 -.477 0.833 -1.195 0.093 -.320 Page 16
Observed and Predicted Frequencies Exp Ed. Level Career Progress Frequency Percentage Pearson Residual Observed Between 1 to 10 years Master Degree Below the expectations Neutral According to the expectations Above the expectations PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations Between 10 to 15 years Bachelor Student Below the expectations Neutral According to the expectations Above the expectations Bachelor Degree Below the expectations Neutral According to the expectations Above the expectations Master Student Below the expectations Neutral According to the expectations Above the expectations Master Degree Below the expectations Neutral According to the expectations Above the expectations PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations 1.287 50.0% 22.9% -.786 0.0% 13.4% .190 50.0% 45.3% -.950 0.0% 18.4% -.658 0.0% 12.6% 1.690 66.7% 24.6% -1.146 0.0% 30.4% .037 33.3% 32.3% -.691 0.0% 32.3% -.639 0.0% 29.0% 2.513 100.0% 13.7% -.578 0.0% 25.0% -.310 0.0% 8.7% -.307 0.0% 8.6% .957 100.0% 52.2% -.661 0.0% 30.4% .364 21.2% 18.7% -.008 9.1% 9.1% -.137 42.4% 43.6% -.158 27.3% 28.5% -1.095 0.0% 28.6% 1.477 33.3% 9.0% .093 33.3% 30.8% .064 33.3% 31.6% -.284 11.1% 14.4% -.324 11.1% 15.0% -.229 33.3% 37.0% .691 44.4% 33.6% -.462 0.0% 6.6% -.949 0.0% 23.1% .532 33.3% 20.9% .598 66.7% 49.4% 1.143 50.0% 18.6% .634 50.0% 29.6% -.477 0.0% 10.2% -1.195 0.0% 41.7% -.320 0.0% 9.3% Page 17
Observed and Predicted Frequencies Exp Ed. Level Career Progress Percentage Predicted Between 1 to 10 years Master Degree Below the expectations Neutral According to the expectations Above the expectations PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations Between 10 to 15 years Bachelor Student Below the expectations Neutral According to the expectations Above the expectations Bachelor Degree Below the expectations Neutral According to the expectations Above the expectations Master Student Below the expectations Neutral According to the expectations Above the expectations Master Degree Below the expectations Neutral According to the expectations Above the expectations PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations 22.9% 13.4% 45.3% 18.4% 12.6% 24.6% 30.4% 32.3% 32.3% 29.0% 13.7% 25.0% 8.7% 8.6% 52.2% 30.4% 18.7% 9.1% 43.6% 28.5% 28.6% 9.0% 30.8% 31.6% 14.4% 15.0% 37.0% 33.6% 6.6% 23.1% 20.9% 49.4% 18.6% 29.6% 10.2% 41.7% 9.3% Page 18
Observed and Predicted Frequencies Exp Ed. Level Career Progress Frequency Observed Predicted More than 15 years Bachelor Student Below the expectations Neutral According to the expectations Above the expectations Bachelor Degree Below the expectations Neutral According to the expectations Above the expectations Master Student Below the expectations Neutral According to the expectations Above the expectations Master Degree Below the expectations Neutral According to the expectations Above the expectations PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations 0.093 -.320 0.035 -.192 1.610 .800 0.262 -.595 57.220 -.924 21.364 .555 20 18.507 .498 98.909 .035 52.493 1.913 0.304 -.562 12.966 -1.439 22.238 -.187 22.782 -.513 21.118 .864 87.857 .069 55.243 -.128 1.423 .928 0.569 -.801 21.461 .530 22.547 -.490 1.989 .012 0.611 -.849 0.599 -.840 31.801 1.205 Page 19
Observed and Predicted Frequencies Exp Ed. Level Career Progress Frequency Percentage Pearson Residual Observed More than 15 years Bachelor Student Below the expectations Neutral According to the expectations Above the expectations Bachelor Degree Below the expectations Neutral According to the expectations Above the expectations Master Student Below the expectations Neutral According to the expectations Above the expectations Master Degree Below the expectations Neutral According to the expectations Above the expectations PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations -.320 0.0% 9.3% -.192 0.0% 3.5% .800 100.0% 61.0% -.595 0.0% 26.2% -.924 13.9% 20.1% .555 5.6% 3.8% .498 55.6% 51.4% .035 25.0% 24.7% 1.913 62.5% 31.2% -.562 0.0% 3.8% -1.439 12.5% 37.1% -.187 25.0% 28.0% -.513 11.8% 16.4% .864 11.8% 6.6% .069 47.1% 46.2% -.128 29.4% 30.8% .928 20.0% 8.5% -.801 0.0% 11.4% .530 40.0% 29.2% -.490 40.0% 50.9% .012 25.0% 24.7% -.849 0.0% 15.3% -.840 0.0% 15.0% 1.205 75.0% 45.0% Page 20
Observed and Predicted Frequencies Exp Ed. Level Career Progress Percentage Predicted More than 15 years Bachelor Student Below the expectations Neutral According to the expectations Above the expectations Bachelor Degree Below the expectations Neutral According to the expectations Above the expectations Master Student Below the expectations Neutral According to the expectations Above the expectations Master Degree Below the expectations Neutral According to the expectations Above the expectations PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations 9.3% 3.5% 61.0% 26.2% 20.1% 3.8% 51.4% 24.7% 31.2% 3.8% 37.1% 28.0% 16.4% 6.6% 46.2% 30.8% 8.5% 11.4% 29.2% 50.9% 24.7% 15.3% 15.0% 45.0% The percentages are based on total observed frequencies in each subpopulation. VIF - Career Progress - > HC (men) Page 21
Notes Output Created Comments Input Data Active Dataset Filter Weight Split File N of Rows in Working Data File Missing Value Handling Definition of Missing Cases Used Syntax Resources Processor Time Elapsed Time Memory Required Additional Memory Required for Residual Plots 30-JAN-2025 19:55:57 /Users/mendesf1/Librar y/CloudStorage/OneDriv eAaltoUniversity/Working Papers/IST - Subjective Career Success - ... DataSet2 <none> <none> <none> 176 User-defined missing values are treated as missing. Statistics are based on cases with no missing values for any variable used. REGRESSION /MISSING LISTWISE /STATISTICS COLLIN TOL /CRITERIA=PIN(.05) POUT(.10) TOLERANCE(. 0001) /NOORIGIN /DEPENDENT CareerProg /METHOD=ENTER Ed. Level Exp. 00:00:00.01 00:00:00.00 3248 bytes 0 bytes Variables Entered/Removeda Model Variables Entered Variables Removed Method 1 Exp, Ed. Levelb.Enter Dependent Variable: Career Progressa. All requested variables entered.b. Page 22
Coefficientsa Model Collinearity Statistics Tolerance VIF 1Ed. Level Exp .807 1.239 .807 1.239 Dependent Variable: Career Progress a. Nominal Regression - Career Progress - > HC (men) Notes Output Created Comments Input Data Active Dataset Filter Weight Split File N of Rows in Working Data File Missing Value Handling Definition of Missing Cases Used Syntax 30-JAN-2025 19:53:52 /Users/mendesf1/Librar y/CloudStorage/OneDriv eAaltoUniversity/Working Papers/IST - Subjective Career Success - ... DataSet2 <none> <none> <none> 176 User-defined missing values are treated as missing. Statistics are based on all cases with valid data for all variables in the model. NOMREG CareerProg (BASE=FIRST ORDER=ASCENDING) BY Ed.Level Exp /CRITERIA CIN(95) DELTA(0) MXITER(100) MXSTEP(5) CHKSEP(20) LCONVERGE(0) PCONVERGE(0.000001) SINGULAR (0.00000001) /MODEL /STEPWISE=PIN(.05) POUT(0.1) MINEFFECT(0) RULE(SINGLE) ENTRYMETHOD(LR) REMOVALMETHOD(LR) /INTERCEPT=INCLUDE /PRINT=CELLPROB PARAMETER SUMMARY LRT CPS STEP MFI. 00:00:00.04 Page 23
Notes Resources Processor Time Elapsed Time 00:00:00.04 00:00:00.00 Warnings Unexpected singularities in the Hessian matrix are encountered. This indicates that either some predictor variables should be excluded or some categories should be merged. The NOMREG procedure continues despite the above warning(s). Subsequent results shown are based on the last iteration. Validity of the model fit is uncertain. Case Processing Summary NMarginal Percentage Career Progress Below the expectations Neutral According to the expectations Above the expectations Ed. Level Bachelor Student Bachelor Degree Master Student Master Degree PhD Student PhD Degree Exp Less than 1 year Between 1 to 10 years Between 10 to 15 years More than 15 years Valid Missing Total Subpopulation 21 11.9% 22 12.5% 74 42.0% 59 33.5% 44 25.0% 78 44.3% 14 8.0% 23 13.1% 10 5.7% 74.0% 16 9.1% 78 44.3% 35 19.9% 47 26.7% 176 100.0% 0 176 20a The dependent variable has only one value observed in 7 (35.0%) subpopulations. a. Model Fitting Information Model Model Fitting Criteria Likelihood Ratio Tests -2 Log Likelihood Chi-Square df Sig. Intercept Only Final 151.802 100.667 51.134 24 .001 Page 24
Pseudo R-Square Cox and Snell Nagelkerke McFadden .252 .275 .117 Likelihood Ratio Tests Effect Model Fitting Criteria Likelihood Ratio Tests -2 Log Likelihood of Reduced Model Chi-Square df Sig. Intercept Ed. Level Exp 100.667a.000 0 . 131.950 31.283 15 .008 126.410 25.742 9.002 The chi-square statistic is the difference in -2 loglikelihoods between the final model and a reduced model. The reduced model is formed by omitting an effect from the final model. The null hypothesis is that all parameters of that effect are 0. This reduced model is equivalent to the final model because omitting the effect does not increase the degrees of freedom. a. Parameter Estimates Career Progressa BStd. Error Wald df Neutral Intercept [Ed. Level=2] [Ed. Level=3] [Ed. Level=4] [Ed. Level=5] [Ed. Level=6] [Ed. Level=7] [Exp=1] [Exp=2] [Exp=3] [Exp=4] According to the expectations Intercept [Ed. Level=2] [Ed. Level=3] [Ed. Level=4] [Ed. Level=5] [Ed. Level=6] [Ed. Level=7] [Exp=1] [Exp=2] .930 1.514 .377 1.539 -.069 1.519 .002 1.964 -.902 1.416 .406 1.524 -22.708 .000 . 1 . -.009 1.778 .000 1.996 17.431 1.293 181.694 1<.001 0b. . 0 . -1.153 1.395 .683 1.408 .540 1.067 .256 1.613 .208 1.167 .032 1.858 0b. . 0 . 1.101 1.647 .446 1.504 3.121 1.761 3.140 1.076 1.342 1.626 .680 1.409 -1.029 1.713 .361 1.548 1.641 1.896 .749 1.387 18.542 1.372 182.764 1<.001 0b. . 0 . -5.041 1.374 13.453 1<.001 -.868 .820 1.118 1.290 -.948 .957 .982 1.322 Page 25
Observed and Predicted Frequencies Exp Ed. Level Career Progress Frequency Observed Predicted Between 10 to 15 years Bachelor Degree Below the expectations Neutral According to the expectations Above the expectations Master Student Below the expectations Neutral According to the expectations Above the expectations Master Degree Below the expectations Neutral According to the expectations Above the expectations PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations More than 15 years Bachelor Student Below the expectations Neutral According to the expectations Above the expectations Bachelor Degree Below the expectations Neutral According to the expectations Above the expectations Master Student Below the expectations Neutral According to the expectations Above the expectations Master Degree Below the expectations Neutral According to the expectations Above the expectations 21.799 .157 32.278 .507 78.011 -.454 98.912 .039 0.531 -1.064 0.000 .000 1.221 1.877 0.248 -.574 0.466 -.704 11.443 -.407 32.803 .146 43.288 .511 0.000 .000 0.642 -.904 1.728 .367 21.630 .429 1.204 1.859 1.637 .551 0.238 -.519 0.921 -1.307 0.010 -.101 0.024 -.157 1.689 .671 0.277 -.618 0.989 -1.017 11.017 -.017 13 11.366 .697 88.628 -.274 21.381 .651 0.000 .000 11.483 -.500 11.136 -.151 1.413 .927 21.039 .984 56.413 -.784 55.135 -.076 0.000 .000 Page 32
Observed and Predicted Frequencies Exp Ed. Level Career Progress Frequency Percentage Pearson Residual Observed Between 10 to 15 years Bachelor Degree Below the expectations Neutral According to the expectations Above the expectations Master Student Below the expectations Neutral According to the expectations Above the expectations Master Degree Below the expectations Neutral According to the expectations Above the expectations PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations More than 15 years Bachelor Student Below the expectations Neutral According to the expectations Above the expectations Bachelor Degree Below the expectations Neutral According to the expectations Above the expectations Master Student Below the expectations Neutral According to the expectations Above the expectations Master Degree Below the expectations Neutral According to the expectations Above the expectations .157 9.5% 8.6% .507 14.3% 10.8% -.454 33.3% 38.1% .039 42.9% 42.4% -1.064 0.0% 53.1% .000 0.0% 0.0% 1.877 100.0% 22.1% -.574 0.0% 24.8% -.704 0.0% 5.8% -.407 12.5% 18.0% .146 37.5% 35.0% .511 50.0% 41.1% .000 0.0% 0.0% -.904 0.0% 21.4% .367 33.3% 24.3% .429 66.7% 54.3% 1.859 50.0% 10.2% .551 50.0% 31.9% -.519 0.0% 11.9% -1.307 0.0% 46.0% -.101 0.0% 1.0% -.157 0.0% 2.4% .671 100.0% 68.9% -.618 0.0% 27.7% -1.017 0.0% 4.5% -.017 4.5% 4.6% .697 59.1% 51.7% -.274 36.4% 39.2% .651 50.0% 34.5% .000 0.0% 0.0% -.500 25.0% 37.1% -.151 25.0% 28.4% .927 7.7% 3.2% .984 15.4% 8.0% -.784 38.5% 49.3% -.076 38.5% 39.5% .000 0.0% 0.0% Page 33
Observed and Predicted Frequencies Exp Ed. Level Career Progress Percentage Predicted Between 10 to 15 years Bachelor Degree Below the expectations Neutral According to the expectations Above the expectations Master Student Below the expectations Neutral According to the expectations Above the expectations Master Degree Below the expectations Neutral According to the expectations Above the expectations PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations More than 15 years Bachelor Student Below the expectations Neutral According to the expectations Above the expectations Bachelor Degree Below the expectations Neutral According to the expectations Above the expectations Master Student Below the expectations Neutral According to the expectations Above the expectations Master Degree Below the expectations Neutral According to the expectations Above the expectations 8.6% 10.8% 38.1% 42.4% 53.1% 0.0% 22.1% 24.8% 5.8% 18.0% 35.0% 41.1% 0.0% 21.4% 24.3% 54.3% 10.2% 31.9% 11.9% 46.0% 1.0% 2.4% 68.9% 27.7% 4.5% 4.6% 51.7% 39.2% 34.5% 0.0% 37.1% 28.4% 3.2% 8.0% 49.3% 39.5% 0.0% Page 34
Observed and Predicted Frequencies Exp Ed. Level Career Progress Frequency Observed Predicted More than 15 years PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations 0.000 .000 0.396 -.663 21.425 .600 22.179 -.180 0.207 -.472 0.525 -.798 0.622 -.886 31.646 1.571 Observed and Predicted Frequencies Exp Ed. Level Career Progress Frequency Percentage Pearson Residual Observed More than 15 years PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations .000 0.0% 0.0% -.663 0.0% 9.9% .600 50.0% 35.6% -.180 50.0% 54.5% -.472 0.0% 6.9% -.798 0.0% 17.5% -.886 0.0% 20.7% 1.571 100.0% 54.9% Observed and Predicted Frequencies Exp Ed. Level Career Progress Percentage Predicted More than 15 years PhD Student Below the expectations Neutral According to the expectations Above the expectations PhD Degree Below the expectations Neutral According to the expectations Above the expectations 0.0% 9.9% 35.6% 54.5% 6.9% 17.5% 20.7% 54.9% The percentages are based on total observed frequencies in each subpopulation. VIF - Career Progress - > HC (women) Page 35
Notes Output Created Comments Input Data Active Dataset Filter Weight Split File N of Rows in Working Data File Missing Value Handling Definition of Missing Cases Used Syntax Resources Processor Time Elapsed Time Memory Required Additional Memory Required for Residual Plots 30-JAN-2025 20:06:13 /Users/mendesf1/Librar y/CloudStorage/OneDriv eAaltoUniversity/Working Papers/IST - Subjective Career Success - ... DataSet3 <none> <none> <none> 155 User-defined missing values are treated as missing. Statistics are based on cases with no missing values for any variable used. REGRESSION /MISSING LISTWISE /STATISTICS COLLIN TOL /CRITERIA=PIN(.05) POUT(.10) TOLERANCE(. 0001) /NOORIGIN /DEPENDENT CareerProg /METHOD=ENTER Ed. Level Exp. 00:00:00.01 00:00:00.00 3248 bytes 0 bytes Variables Entered/Removeda Model Variables Entered Variables Removed Method 1 Exp, Ed. Levelb.Enter Dependent Variable: Career Progressa. All requested variables entered.b. Page 36
Coefficientsa Model Collinearity Statistics Tolerance VIF 1Ed. Level Exp .765 1.307 .765 1.307 Dependent Variable: Career Progress a. Nominal Regression - Career Progress - > HC (women) Warnings Unexpected singularities in the Hessian matrix are encountered. This indicates that either some predictor variables should be excluded or some categories should be merged. The NOMREG procedure continues despite the above warning(s). Subsequent results shown are based on the last iteration. Validity of the model fit is uncertain. Case Processing Summary NMarginal Percentage Career Progress Below the expectations Neutral According to the expectations Above the expectations Ed. Level High School Bachelor Student Bachelor Degree Master Student Master Degree PhD Student PhD Degree Exp Less than 1 year Between 1 to 10 years Between 10 to 15 years More than 15 years Valid Missing Total Subpopulation 54 34.8% 95.8% 82 52.9% 10 6.5% 53.2% 46 29.7% 80 51.6% 15 9.7% 74.5% 10.6% 10.6% 18 11.6% 97 62.6% 16 10.3% 24 15.5% 155 100.0% 0 155 18a The dependent variable has only one value observed in 7 (38.9%) subpopulations. a. Page 37
Model Fitting Information Model Model Fitting Criteria Likelihood Ratio Tests -2 Log Likelihood Chi-Square df Sig. Intercept Only Final 107.187 76.625 30.563 27 .289 Pseudo R-Square Cox and Snell Nagelkerke McFadden .179 .204 .094 Likelihood Ratio Tests Effect Model Fitting Criteria Likelihood Ratio Tests -2 Log Likelihood of Reduced Model Chi-Square df Sig. Intercept Ed. Level Exp 76.625a.000 0 . 98.917 22.292 18 .219 84.290 7.665 9.568 The chi-square statistic is the difference in -2 loglikelihoods between the final model and a reduced model. The reduced model is formed by omitting an effect from the final model. The null hypothesis is that all parameters of that effect are 0. This reduced model is equivalent to the final model because omitting the effect does not increase the degrees of freedom. a. Page 38
Parameter Estimates Career Progressa BStd. Error Wald df Neutral Intercept [Ed. Level=1] [Ed. Level=2] [Ed. Level=3] [Ed. Level=4] [Ed. Level=5] [Ed. Level=6] [Ed. Level=7] [Exp=1] [Exp=2] [Exp=3] [Exp=4] According to the expectations Intercept [Ed. Level=1] [Ed. Level=2] [Ed. Level=3] [Ed. Level=4] [Ed. Level=5] [Ed. Level=6] [Ed. Level=7] [Exp=1] [Exp=2] [Exp=3] [Exp=4] Above the expectations Intercept [Ed. Level=1] [Ed. Level=2] [Ed. Level=3] [Ed. Level=4] [Ed. Level=5] [Ed. Level=6] [Ed. Level=7] [Exp=1] [Exp=2] [Exp=3] [Exp=4] -18.823 8645.867 .000 1.998 .641 9196.370 .000 11.000 17.760 8645.868 .000 1.998 16.429 8645.868 .000 1.998 18.095 8645.868 .000 1.998 1.422 9169.585 .000 11.000 .000 .000 . 1 . 0c. . 0 . 1.605 1.489 1.162 1.281 -.339 1.275 .071 1.790 .293 1.553 .036 1.850 0c. . 0 . -16.613 4050.773 .000 1.997 15.114 4050.773 .000 1.997 17.657 4050.773 .000 1.997 16.808 4050.773 .000 1.997 16.945 4050.773 .000 1.997 16.303 4050.773 .000 1.997 .000 5728.658 .000 11.000 0c. . 0 . .703 .860 .669 1.413 .043 .554 .006 1.937 .073 .723 .010 1.919 0c. . 0 . -18.717 8202.191 .000 1.998 18.851 8202.191 .000 1.998 18.369 8202.191 .000 1.998 16.823 8202.191 .000 1.998 18.857 8202.191 .000 1.998 1.600 8688.365 .000 11.000 .000 .000 . 1 . 0c. . 0 . -15.689 1528.576 .000 1.992 -.617 1.041 .351 1.553 -.362 1.400 .067 1.796 0c. . 0 . Page 39
Parameter Estimates Career Progressa Sig. Exp(B) 95% Confidence ... Lower Bound Neutral Intercept [Ed. Level=1] [Ed. Level=2] [Ed. Level=3] [Ed. Level=4] [Ed. Level=5] [Ed. Level=6] [Ed. Level=7] [Exp=1] [Exp=2] [Exp=3] [Exp=4] According to the expectations Intercept [Ed. Level=1] [Ed. Level=2] [Ed. Level=3] [Ed. Level=4] [Ed. Level=5] [Ed. Level=6] [Ed. Level=7] [Exp=1] [Exp=2] [Exp=3] [Exp=4] Above the expectations Intercept [Ed. Level=1] [Ed. Level=2] [Ed. Level=3] [Ed. Level=4] [Ed. Level=5] [Ed. Level=6] [Ed. Level=7] [Exp=1] [Exp=2] [Exp=3] [Exp=4] .998 1.000 1.899 .000 .b .998 51640031.7 .000 .b .998 13644749.1 .000 .b .998 72225406.2 .000 .b 1.000 4.146 .000 .b .1.000 1.000 1.000 .... .281 4.978 .269 92.105 .790 .713 .059 8.672 .850 1.341 .064 28.153 .... .997 .997 3663793.355 .000 .b .997 46616003.2 .000 .b .997 19937870.3 .000 .b .997 22861716.2 .000 .b .997 12027680.0 .000 .b 1.000 1.000 .000 .b .... .413 2.021 .374 10.903 .937 1.044 .353 3.092 .919 1.076 .261 4.437 .... .998 .998 153793672 .000 .b .998 95006203.6 .000 .b .998 20243446.6 .000 .b .998 154770528 .000 .b 1.000 4.952 .000 .b .1.000 1.000 1.000 .... .992 1.535E-7 .000 .b .553 .539 .070 4.151 .796 .696 .045 10.832 .... Page 40
Parameter Estimates Career Progressa 95% Confidence Interval for Exp... Upper Bound Neutral Intercept [Ed. Level=1] [Ed. Level=2] [Ed. Level=3] [Ed. Level=4] [Ed. Level=5] [Ed. Level=6] [Ed. Level=7] [Exp=1] [Exp=2] [Exp=3] [Exp=4] According to the expectations Intercept [Ed. Level=1] [Ed. Level=2] [Ed. Level=3] [Ed. Level=4] [Ed. Level=5] [Ed. Level=6] [Ed. Level=7] [Exp=1] [Exp=2] [Exp=3] [Exp=4] Above the expectations Intercept [Ed. Level=1] [Ed. Level=2] [Ed. Level=3] [Ed. Level=4] [Ed. Level=5] [Ed. Level=6] [Ed. Level=7] [Exp=1] [Exp=2] [Exp=3] [Exp=4] .b .b .b .b .b 1.000 . 92.105 8.672 28.153 . .b .b .b .b .b .b . 10.903 3.092 4.437 . .b .b .b .b .b 1.000 . .b 4.151 10.832 . The reference category is: Below the expectations.a. Floating point overflow occurred while computing this statistic. Its value is therefore set to system missing.b. This parameter is set to zero because it is redundant.c. Page 41
Warnings There are 1 (12.5%) cells (i.e., dependent variable levels by subpopulations) with zero frequencies. Unexpected singularities in the Hessian matrix are encountered. This indicates that either some predictor variables should be excluded or some categories should be merged. The NOMREG procedure continues despite the above warning(s). Subsequent results shown are based on the last iteration. Validity of the model fit is uncertain. Measures of Monotone Association table is not generated because the dependent variable does not have exactly two levels. Case Processing Summary NMarginal Percentage Career Progress Below the expectations Neutral According to the expectations Above the expectations None FALSE TRUE Valid Missing Total Subpopulation 54 34.8% 95.8% 82 52.9% 10 6.5% 125 80.6% 30 19.4% 155 100.0% 0 155 2 Model Fitting Information Model Model Fitting Criteria Likelihood Ratio Tests AIC BIC -2 Log Likelihood Chi-Square df Sig. Intercept Only Final 43.132 52.263 37.132 30.378 48.639 18.378 18.754 3<.001 Goodness-of-Fit Chi-Square df Sig. Pearson Deviance .000 0 . .000 0 . Pseudo R-Square Cox and Snell Nagelkerke McFadden .114 .130 .058 Page 48
Likelihood Ratio Tests Effect Model Fitting Criteria Likelihood Ratio Tests AIC of Reduced Model BIC of Reduced Model -2 Log Likelihood of Reduced Model Chi-Square df Sig. Intercept None 30.378 48.639 18.378a.000 0 . 43.132 52.263 37.132 18.754 3<.001 The chi-square statistic is the difference in -2 log-likelihoods between the final model and a reduced model. The reduced model is formed by omitting an effect from the final model. The null hypothesis is that all parameters of that effect are 0. This reduced model is equivalent to the final model because omitting the effect does not increase the degrees of freedom. a. Parameter Estimates Career Progressa BStd. Error Wald df Sig. Neutral Intercept [None=0] [None=1] According to the expectations Intercept [None=0] [None=1] Above the expectations Intercept [None=0] [None=1] -.405 .913 .197 1.657 -1.580 .998 2.508 1.113 .206 0b. . 0 . . 2.120 .611 12.042 1<.001 -2.009 .641 9.833 1.002 .134 0b. . 0 . . -17.642 .346 2602.260 1<.001 16.013 .000 . 1 . 9002909.349 0b. . 0 . . Parameter Estimates Career Progressa Exp(B) 95% Confidence Interval for Exp (B) Lower Bound Upper Bound Neutral Intercept [None=0] [None=1] According to the expectations Intercept [None=0] [None=1] Above the expectations Intercept [None=0] [None=1] .206 .029 1.456 ... .134 .038 .471 ... 9002909.349 9002909.349 9002909.349 ... The reference category is: Below the expectations.a. This parameter is set to zero because it is redundant.b. Page 49
Classification Observed Predicted Below the expectations Neutral According to the expectations Above the expectations Below the expectations Neutral According to the expectations Above the expectations Overall Percentage 0 0 54 00.0% 0 0 9 0 0.0% 0 0 82 0100.0% 0 0 10 00.0% 0.0% 0.0% 100.0% 0.0% 52.9% Classification Observed Predicted Percent Correct Below the expectations Neutral According to the expectations Above the expectations Overall Percentage 0.0% 0.0% 100.0% 0.0% 52.9% Observed and Predicted Frequencies None Career Progress Frequency Percentage Observed Predicted Pearson Residual Observed Predicted FALSE Below the expectations Neutral According to the expectations Above the expectations TRUE Below the expectations Neutral According to the expectations Above the expectations 51 51.000 .000 40.8% 40.8% 77.000 .000 5.6% 5.6% 57 57.000 .000 45.6% 45.6% 10 10.000 .000 8.0% 8.0% 33.000 .000 10.0% 10.0% 22.000 .000 6.7% 6.7% 25 25.000 .000 83.3% 83.3% 0.000 .000 0.0% 0.0% The percentages are based on total observed frequencies in each subpopulation. Page 50