1 OLS REGRESSION

OLS=lm(WP.A.PER~WP.PG.AST+ WP.DefensePerformance+ WT.FI.W+ WT.BS.TP.Age+ WT.NBAExperience.Mean+ WT.TimeCleanOrder.SD+ WT.PerGame.GamesStarted.SD+ WT.BS.TS.AST+ WP.NBA.1stYear+ WP.PG.PF+ WP.PG.G+ BP.Height.Cm+ BP.Weight.Kg+ WP.Age, data=Sdata)
#VIF>5: WP.PG.PTS+ WP.PG.MP+ 
#print_table(OLS)
#tab_model(OLS)
GLM_summary(OLS)
## 
## General Linear Model (OLS Regression)
## 
## Model Fit:
## F(14, 527) = 27.97, p = 8e-55 ***
## R² = 0.42627 (Adjusted R² = 0.41103)
## 
## Unstandardized Coefficients:
## Outcome Variable: WP.A.PER
## N = 542 (168 missing cases deleted)
## ──────────────────────────────────────────────────────────────────────────────────
##                                  b    S.E.      t     p        [95% CI of b]   VIF
## ──────────────────────────────────────────────────────────────────────────────────
## (Intercept)                  0.000 (0.034)  0.012  .990     [-0.067,  0.068]      
## WP.PG.AST                    0.275 (0.048)  5.753 <.001 *** [ 0.181,  0.369] 2.091
## WP.DefensePerformance        0.527 (0.059)  8.885 <.001 *** [ 0.411,  0.644] 3.134
## WT.FI.W                      0.083 (0.054)  1.541  .124     [-0.023,  0.188] 2.552
## WT.BS.TP.Age                -0.063 (0.054) -1.164  .245     [-0.170,  0.044] 2.596
## WT.NBAExperience.Mean       -0.023 (0.046) -0.508  .612     [-0.114,  0.067] 2.072
## WT.TimeCleanOrder.SD         0.033 (0.042)  0.799  .425     [-0.048,  0.115] 1.682
## WT.PerGame.GamesStarted.SD  -0.062 (0.045) -1.379  .169     [-0.149,  0.026] 1.578
## WT.BS.TS.AST                 0.002 (0.040)  0.057  .955     [-0.076,  0.081] 1.435
## WP.NBA.1stYear              -0.022 (0.040) -0.545  .586     [-0.099,  0.056] 1.244
## WP.PG.PF                    -0.198 (0.051) -3.848 <.001 *** [-0.299, -0.097] 2.359
## WP.PG.G                      0.119 (0.041)  2.879  .004 **  [ 0.038,  0.201] 1.522
## BP.Height.Cm                -0.009 (0.064) -0.132  .895     [-0.135,  0.118] 3.696
## BP.Weight.Kg                 0.121 (0.062)  1.963  .050 .   [-0.000,  0.243] 3.273
## WP.Age                      -0.034 (0.038) -0.906  .365     [-0.109,  0.040] 1.211
## ──────────────────────────────────────────────────────────────────────────────────
## 
## Standardized Coefficients (β):
## Outcome Variable: WP.A.PER
## N = 542 (168 missing cases deleted)
## ───────────────────────────────────────────────────────────────────────────────────────────────
##                                  β    S.E.      t     p        [95% CI of β] r(partial) r(part)
## ───────────────────────────────────────────────────────────────────────────────────────────────
## WP.PG.AST                    0.274 (0.048)  5.753 <.001 *** [ 0.181,  0.368]      0.243   0.190
## WP.DefensePerformance        0.519 (0.058)  8.885 <.001 *** [ 0.404,  0.634]      0.361   0.293
## WT.FI.W                      0.081 (0.053)  1.541  .124     [-0.022,  0.185]      0.067   0.051
## WT.BS.TP.Age                -0.062 (0.053) -1.164  .245     [-0.166,  0.043]     -0.051  -0.038
## WT.NBAExperience.Mean       -0.024 (0.047) -0.508  .612     [-0.117,  0.069]     -0.022  -0.017
## WT.TimeCleanOrder.SD         0.034 (0.043)  0.799  .425     [-0.050,  0.118]      0.035   0.026
## WT.PerGame.GamesStarted.SD  -0.057 (0.041) -1.379  .169     [-0.139,  0.024]     -0.060  -0.045
## WT.BS.TS.AST                 0.002 (0.040)  0.057  .955     [-0.075,  0.080]      0.002   0.002
## WP.NBA.1stYear              -0.020 (0.037) -0.545  .586     [-0.092,  0.052]     -0.024  -0.018
## WP.PG.PF                    -0.195 (0.051) -3.848 <.001 *** [-0.295, -0.095]     -0.165  -0.127
## WP.PG.G                      0.117 (0.041)  2.879  .004 **  [ 0.037,  0.197]      0.124   0.095
## BP.Height.Cm                -0.008 (0.063) -0.132  .895     [-0.133,  0.116]     -0.006  -0.004
## BP.Weight.Kg                 0.117 (0.060)  1.963  .050 .   [-0.000,  0.234]      0.085   0.065
## WP.Age                      -0.033 (0.036) -0.906  .365     [-0.104,  0.038]     -0.039  -0.030
## ───────────────────────────────────────────────────────────────────────────────────────────────
#OLS=lm(WP.A.PER~WT.NBAExperience.Mean+ WT.TimeCleanOrder.SD+ WT.PerGame.GamesStarted.SD+ WP.NBA.1stYear+ BP.Height.Cm+ BP.Weight.Kg+ WP.Age, data=Sdata)
#OLS=lm(WP.A.PER~WP.PG.PTS+WT.NBAExperience.Mean+ WT.TimeCleanOrder.SD+ WT.PerGame.GamesStarted.SD+ WP.NBA.1stYear+ BP.Height.Cm+ BP.Weight.Kg+ WP.Age, data=Sdata)
tab_model(OLS,show.se = TRUE, show.p = TRUE, show.df = F, show.r2 = TRUE, show.re.var = TRUE, show.aic = TRUE, show.aicc = TRUE, show.dev = TRUE, show.loglik = TRUE, show.obs = TRUE, show.reflvl = TRUE,digits = 3,title="Table 7. Results of OLS regression for job performance",file = "OLS.doc")
Table 7. Results of OLS regression for job performance
  Job performance
Predictors Estimates std. Error CI p
(Intercept) 0.000 0.034 -0.067 – 0.068 0.990
Coworker-serving behavior 0.275 0.048 0.181 – 0.369 <0.001
Team-serving behavior 0.527 0.059 0.411 – 0.644 <0.001
Team performance 0.083 0.054 -0.023 – 0.188 0.124
Average age of all
players in a team 		-0.063 0.054 -0.170 – 0.044 0.245
Average tenure of team
member 		-0.023 0.046 -0.114 – 0.067 0.612
Inequality of team
members’ playing time 		0.033 0.042 -0.048 – 0.115 0.425
Inequality of the number
of starts 		-0.062 0.045 -0.149 – 0.026 0.169
Team cooperation 0.002 0.040 -0.076 – 0.081 0.955
Age of joining the team -0.022 0.040 -0.099 – 0.056 0.586
Uncooperative behavior -0.198 0.051 -0.299 – -0.097 <0.001
Working time 0.119 0.041 0.038 – 0.201 0.004
Height -0.009 0.064 -0.135 – 0.118 0.895
Weight 0.121 0.062 -0.000 – 0.243 0.050
Player age -0.034 0.038 -0.109 – 0.040 0.365
Observations 542
R2 / R2 adjusted 0.426 / 0.411
Deviance 326.096
AIC 1294.753
AICc 1295.789
log-Likelihood -631.377 
#Corr(data)
#GLM_summary(OLS)

2 PLOT

plot_model(OLS, sort.est = TRUE)

3 REPORT

For statistical modeling, we applied OLS regression to examine all predictors of job performance in ML. While the model’s overall performance of explainability does not surpass that ofSAFL, it exceeds ALMMo and several ML (e.g., LASSO, BRR, RF), with R² valued at 0.528 and an adjusted R² of 0.513. This indicates the model based on OLS regression, boasting an R² of 0.528 and an adjusted R² of 0.513, significantly elucidates more than half the variability in job performance. On the significance front, both self-serving and team-serving behaviors were strongly linked to job performance improvements, evidenced by estimates of 1.040 (p < 0.001) and 0.433 (p < 0.001), respectively. Moreover, coworker-serving behavior and the average age of all team players were positively correlated with job performance, as demonstrated by estimates of 0.129 (p = 0.011) and 0.132 (p = 0.017). Noteworthy is the finding that workload Working amount was inversely related to job performance, with an estimate of -1.036 (p < 0.001), suggesting a detrimental effect of increased workload on performance. Other variables did not exhibit statistical significance in this context. This outcome underscores that XML slightly outperforms ordinary least squares (OLS) regression in performance of explainability. The enhanced data fit provided by XML can be ascribed to its methodology of establishing multiple local models, thereby capturing the nuances within the dataset more effectively.

#report(OLS)