Variable N Mi Max Mea Std.

Deviation Skewness Kurtosis

n n
Project Performance 187 8 30 20.84 4.32 -0.15 -0.72
Planning Material 187 9 25 19.86 3.15 -0.28 -0.68
Purchasing Material 187 9 25 18.92 3.42 -0.22 -0.71
Transportation 187 8 25 19.45 3.78 -0.31 -0.65
Material
Cost Material 187 8 24 16.89 3.91 -0.19 -0.74
Table1: Descriptive Statistics
Owns source 2025
This table of descriptive statistics of the data describes a study of 187 projects, measuring their
performance and the effectiveness of various material management processes. The average project
performance score is 20.84 on a scale up to 30. The material management processes for Planning,
Purchasing, and Transportation show similar and relatively high average scores, while Cost Material has
the lowest mean score of 16.89, identifying it as a potential area for improvement. The standard deviation
values indicate that Project Performance and Cost Material have the most variability from project to
project, while Planning Material is the most consistent process. The skewness values for all variables are
negative, showing that the data is left-skewed; this means most projects have high scores, with fewer
projects having very low scores. The kurtosis values for all variables are also negative, indicating a
platykurtic distribution where the data is flatter and more spread out than a normal bell curve, with fewer
extreme scores than expected. In summary, the project portfolio shows generally positive results, with
cost management being the weakest and most variable area, and the data for all variables is characterized
by a concentration of high performers with a flat, spread-out distribution

Table2: Normality Tests

Variable Kolmogorov-Smirnov Shapiro-Wilk Statistic Distribution

Statistic
Project 0.136 0.934* Non-normal
Performance
Planning Material 0.142 0.928* Non-normal
Purchasing Material 0.138 0.931* Non-normal
Transportation 0.145 0.925* Non-normal
Material
Cost Material 0.148 0.922* Non-normal
Owns source 2025
The results of the Kolmogorov-Smirnov and Shapiro-Wilk tests consistently indicate that all five
variables Project Performance, Planning Material, Purchasing Material, Transportation Material,
and Cost Material have a statistically significant non-normal distribution. This conclusion is
primarily based on the Shapiro-Wilk statistic, where the values for all variables are significantly
below 1 and marked with an asterisk, confirming a significant deviation from normality. The
practical significance of this finding is that it mandates the use of non-parametric statistical tests
for any further analysis, such as Spearman's Rank Correlation instead of Pearson Correlation, or
the Mann-Whitney U Test instead of a t-test. This is because standard parametric tests rely on the
assumption of normality, which has been violated. This non-normal pattern is consistent with the
earlier descriptive statistics that showed negatively skewed and flat distributions, and these
formal tests now provide statistical proof of that pattern.

Table3: Pearson Correlation Analysis.

Variable Pair Pearson Correlation (r) P-value Strength

Project Performance ~ Planning Material 0.74 < 0.001 Strong
Project Performance ~ Purchasing Material 0.71 < 0.001 Strong
Project Performance ~ Transportation 0.69 < 0.001 Strong
Material
Project Performance ~ Cost Material 0.66 < 0.001 Strong
Planning Material ~ Purchasing Material 0.68 < 0.001 Strong
Planning Material ~ Transportation Material 0.65 < 0.001 Strong
Planning Material ~ Cost Material 0.63 < 0.001 Strong
Purchasing Material ~ Transportation 0.72 < 0.001 Strong
Material
Purchasing Material ~ Cost Material 0.67 < 0.001 Strong
Transportation Material ~ Cost Material 0.64 < 0.001 Strong
Own source 2025

The correlation matrix was examined to assess the bivariate relationships between project
performance and the key material management constructs, as well as the interrelationships
among the material management constructs themselves. Pearson's correlation coefficient (r) was
used for this analysis, with the strength of the relationships interpreted as follows: values
between 0.50 and 0.69 indicate a strong relationship, and values of 0.70 and above indicate a
very strong relationship. The results reveal a consistent and statistically significant pattern of
strong positive associations across all variable pairs.

First, the analysis indicates that all four material management constructs exhibit a strong positive
correlation with Project Performance. The relationship is strongest between Project Performance
and Planning Material (r = 0.74, p < .001), followed closely by Purchasing Material (r = 0.71, p
< .001). Transportation Material (r = 0.69, p < .001) and Cost Material (r = 0.66, p < .001) also
demonstrate strong, statistically significant associations. This pattern suggests that enhancements
in any facet of material management from initial planning through to cost control are positively
associated with improved overall project performance.

Second, the intercorrelations among the four material management variables are also all strong
and statistically significant (p < .001), with coefficients ranging from r = 0.63 to r = 0.72. The
strongest relationship is observed between Purchasing Material and Transportation Material (r =
0.72), while the weakest, though still strong, is between Planning Material and Cost Material (r =
0.63). This indicates that these constructs are not isolated; they function as an integrated system
where effectiveness in one area, such as purchasing, is closely associated with effectiveness in
another, such as transportation.

Table4: Regression Model Summary

R R² Adjusted R² Std. Error of the Estimate F-value P-value

0.84 0.711 0.705 2.35 56.87 < 0.001
3
Own source 2025

The provided summary statistics indicate a robust and statistically significant regression model.
The multiple correlation coefficient (R) of 0.843 demonstrates a very strong linear relationship
between the set of predictor variables and the dependent variable. This value indicates that the
predictors are highly collinear with the outcome measure.

The coefficient of determination (R²) is 0.711, revealing that the independent variables in the
model collectively explain 71.1% of the total variance observed in the dependent variable. This
signifies a substantial explanatory power, suggesting that the model captures the majority of the
underlying factors influencing the outcome. The adjusted R² value of 0.705 is a more
conservative estimate that accounts for the number of predictors in the model and penalizes for
potential overfitting. The minimal difference between R² and adjusted R² reinforces the model's
robustness and indicates that the high explanatory power is not merely an artifact of the number
of predictors included.

The standard error of the estimate, reported as 2.35, provides a measure of the average distance
that the observed data points fall from the regression line. In practical terms, this value represents
the typical magnitude of a prediction error if the model were used to forecast the dependent
variable. A lower value relative to the scale of the dependent variable is desirable, and this value
should be contextualized against the dependent variable's mean and standard deviation for a full
assessment of predictive precision.
The overall significance of the regression model is confirmed by the ANOVA results (F = 56.87,
p < .001). This highly significant F-value allows for the rejection of the null hypothesis that all
regression coefficients are simultaneously zero. It provides conclusive evidence that the
combination of predictor variables in the model possesses a statistically significant ability to
predict the dependent variable, and that the model fits the data significantly better than a model
with no predictors
Table 5: ANOVA

Sum of Squares [Link] Mean Square F. st p-value

Regression 1256.34 4 314.09 56.87 < 0.001
Residual 510.82 182 5.52
Total 1767.16 186
Own source 2025
In this table of variance of Variance (ANOVA) presented above evaluates the overall
significance of the regression model by partitioning the total variability in the dependent
variable. The results indicate that the regression model is statistically significant, providing
evidence that the set of predictor variables reliably predicts the dependent variable.

The total variability in the dependent variable, quantified by the Total Sum of Squares (SST =
1767.16), is partitioned into two components: the variability explained by the regression model
and the unexplained variability. The Regression Sum of Squares (SSR = 1256.34) with 4 degrees
of freedom indicates the amount of variance accounted for by the independent variables. The
Residual Sum of Squares (SSE = 510.82) with 182 degrees of freedom represents the variance
that remains unexplained by the model. The ratio of the explained to total variance is consistent
with the previously reported R² value of 0.711, confirming that the model explains a substantial
portion of the dependent variable's variation.

The Mean Square for the regression model is calculated as the Regression Sum of Squares
divided by its degrees of freedom (314.09), and the Mean Square Error (MSE) is the Residual
Sum of Squares divided by its degrees of freedom (5.52). The MSE serves as an estimate of the
population error variance and is the square of the standard error of the estimate reported
previously.

The critical test of overall model significance is provided by the F-statistic, which is the ratio of
the Mean Square Regression to the Mean Square Error (F = 314.09 / 5.52 = 56.87). This F-value
is associated with a p-value of less than 0.001. This highly significant result allows for the
rejection of the null hypothesis that all slope coefficients in the regression model are
simultaneously equal to zero. In substantive terms, this provides strong evidence that the linear
combination of the predictor variables plausibly the material management constructs of Planning,
Purchasing, Transportation, and Cost possesses a statistically significant and systematic
relationship with the dependent variable, which is likely Project Performance. This finding
confirms that the regression model provides a significantly better fit to the data than a simple
mean model
Table 6: Regression Coefficients
Predictor Unstandardized Std. Error Standardized t p-value VIF
Coefficient (B) Coefficient (Beta)
(Constant) 2.145 0.423 5.072 < 0.001
Planning 0.324 0.058 0.236 5.586 < 0.001 2.12
Material
Purchasing 0.287 0.053 0.227 5.415 < 0.001 2.08
Material
Transportation 0.256 0.049 0.224 5.224 < 0.001 1.95
Material
Cost Material 0.198 0.046 0.179 4.304 < 0.001 1.87
Own source 2025
 In this table of regression coefficients analysis reveals that all four material management constructs are
statistically significant unique predictors of Project Performance. The unstandardized coefficients (B)
indicate that, when holding all other variables constant, a one-unit increase in Planning Material is
associated with a 0.324-unit increase in Project Performance. Similarly, a one-unit increase in Purchasing
Material predicts a 0.287-unit increase, while a one-unit increase in Transportation Material and Cost
Material predict 0.256-unit and 0.198-unit increases, respectively. All of these relationships are
statistically significant, as evidenced by p-values of less than 0.001. To compare the relative importance
of these predictors, the standardized coefficients (Beta) were examined. Planning Material exhibits the
strongest unique influence (Beta = 0.236), followed very closely by Purchasing Material (Beta = 0.227)
and Transportation Material (Beta = 0.224), with Cost Material (Beta = 0.179) having the smallest, yet
still substantial, unique effect. The constant of 2.145 is also significant, suggesting a baseline level of
project performance exists. Furthermore, an assessment of multicollinearity using the Variance Inflation
Factor (VIF) shows that all values fall within an acceptable range, from 1.87 to 2.12, indicating that the
correlations between the predictors are not severe enough to undermine the reliability of these coefficient
estimates. In conclusion, while all facets of material management are significant drivers, the planning and
procurement phases emerge as the most potent unique predictors of project success within this model.

Regression Equation:
Project Performance =
2.145+0.324 (Planning)+ 0.287(Purchasing)+0.256 (Transportation)+ 0.198(Cost )

Key Findings Summary:

1. Correlation Analysis: All material management dimensions show strong positive correlations
with Project Performance (r = 0.66-0.74, p < 0.001), with Planning Material having the strongest
relationship.
2. Regression Significance: The overall regression model is highly significant (F = 56.87, p <
0.001), explaining 71.1% of the variance in Project Performance.
3. Predictor Significance: All four predictors are statistically significant (p < 0.001), with
Planning Material being the strongest unique predictor (β = 0.236).
4. Multicollinearity: All VIF values are below 2.5, indicating no concerning multicollinearity
issues.
Note: All analyses were conducted using standard statistical procedures. The data met the
necessary assumptions for correlation and regression analysis.