CHAPTER FOUR

DATA PRESENTATION AND ANALYSIS

Introduction

In this chapter, the study's analysed data is presented, providing the results of the analysis.

This section includes the presentation of descriptive statistics concerning the respondents.

Additionally, preliminary analysis, such as missing value analysis, outlier detection, and

deletion, was conducted. Furthermore, this section reports on the moderating influence of

organisational affective commitment on the relationship between training and work place

environment employee performance.

Response Rate

The response rate of the survey carried out on respondents, are shown below:

Table 4.1: Distribution of Questionnaires and Response Rate

Questionnaire Frequency Rate (%)
Distributed Questionnaires 236 100
Returned Questionnaires 224 95
Unreturned Questionnaires 12 5
Deleted Questionnaires 6 3
Usable Questionnaire 218 92

A total of 236 questionnaires were distributed, of which 224 were retrieved from the

respondents i.e., a total of 95% from the total of 100. 12 questionnaires were not returned i.e.,

5%. 6 questionnaires were deleted because they were outliers i.e., a percentage of 3% making

the total usable number of questionnaire responses to be 218 i.e., a total percentage of 92%.

Preliminary Analysis
Prior to conducting the final analysis, it is necessary to perform a preliminary analysis. For this

study, preliminary analysis was conducted to assess missing values and outliers in the collected

data. Missing data points were addressed through mean substitution, while outliers were
identified using Mahalanobis' distance. Any identified outlier cases were subsequently removed

from the dataset.

Missing Value

Missing data refers to variables without observations or questions without answers. It is widely

acknowledged that missing values pose challenges in quantitative research studies, and most

studies are susceptible to this issue (Bannon, 2015). As such, it is crucial for quantitative

studies to address missing values and ensure that they do not exceed 10 percent (Hair et al.,

2014). Large amounts of missing data in an analysis can potentially led to unreliable findings.

According to Hair et. Al, (2009), missing data points surpassing 10 percent can significantly

compromise the statistical validity of any data analysis. They further suggested that missing

values below 10 percent can be replaced using mean substitution if they occur randomly. In the

present study, the missing values occurred randomly and were replaced using mean

substitution. Descriptive statistics were employed to assess the number of missing values in the

dataset. Out of the total 10,387 data points in the dataset, only 12 were found to be missing.

This indicates that only 5 percent of the data were missing.

Assessment of Outliers

Statistical outliers are data points within a dataset that deviate significantly from the rest of

the observations (Lai, Zha, Zhao, Wang, & Hu, 2021). In multivariate data, an outlier can

differ greatly in terms of the value of a single variable or in terms of the combination of

values across multiple variables. These outliers are also known as abnormalities, discordant,

deviants, or anomalies in the fields of data mining and statistics (Grentzelos, 2020).

In detecting outliers, researchers in this study employed the Mahalanobis distance D 2 method

(Ghorbani, 2019). This approach was chosen due to the limitations of univariate and bivariate

methods in outlier detection (Suboh & Aziz, 2020). The utilisation of the Mahalanobis
distance D2 is particularly advantageous when analyzing a large number of variables and

examining their multivariate nature (Hair et al., 2014). It calculates the distance between each

observation and the mean center of all observations in multidimensional space, resulting in a

single value for each observation regardless of the number of variables involved (Hair et al.,

2014).

Researchers are recommended to use a significant level of 0.001 to identify cases that are

deemed unsuitable for the study. In this study, the Mahalanobis distance (D2) method was

employed to detect outliers using a significance level of 0.001. As a result, six cases were

identified as outliers and subsequently removed from the analysis. Consequently, only 218

cases remained and were utilized for further analysis.

Normality

Normality test was carried out on the basis of individual items measuring the constructs in the

study. The study utilised skewness and kurtosis statistics to assess the normality. Skewness is

the measure of the symmetry of a distribution. While Kurtosis measures how peak or how flat

a distribution is when compared with a normal distribution (Hair et al., 2014). (Tabachnick &

Fidell, 2013). The variables of the study fall within the range of ±1.96, which serves as the

most commonly used threshold (Hair et al., 2014). As a result, the data displayed some

normality.

Normality Test using Skewness and Kurtosis statistics (n= 218)

Construct N Skewness Std. Error Kurtosis Std. Error

(Skewness) (Kurtosis)
Training 218 0.013 0.153 -0.042 0.304
Work place 218 -0.075 0.153 0.089 0.304
environment
Organisational 218 0.057 0.153 -0.128 0.304
effective
commitment
Employee 218 -0.103 0.153 0.154 0.304
performance

Skewness:
Values close to zero (between -0.5 and 0.5) indicate a relatively symmetrical distribution.
Variable 1: Skewness of 0.013 suggests a very slight right skew.
Variable 2: Skewness of -0.075 suggests a very slight left skew.
Variable 3: Skewness of 0.057 suggests a very slight right skew.
Variable 4: Skewness of -0.103 suggests a very slight left skew.
Kurtosis:
Values close to zero (between -1 and 1) indicate a distribution close to normal.
Variable 1: Kurtosis of -0.042 suggests a distribution close to normal.
Variable 2: Kurtosis of 0.089 suggests a distribution close to normal.
Variable 3: Kurtosis of -0.128 suggests a distribution close to normal.
Variable 4: Kurtosis of 0.154 suggests a distribution close to normal.
These values for skewness and kurtosis are within acceptable ranges for assuming normality.
In real-world data, it's rare to have exact zero skewness and kurtosis, but values close to zero
suggest that the data is approximately normal and suitable for parametric statistical tests.
This table shows that all the variables have very small skewness and kurtosis values,
indicating that the data distribution for each variable is fairly symmetrical and close to
normal, which is ideal for many statistical analyses.
Multicolinearity Test
Construct Tolerance VIF
Training => Employee performance 0.450 2.22
Work-life balance => Employee 0.326 3.07
performance
Job satisfaction => Employee 0.408 2.45
performance
Organizational Commitment => 0.515 1.94
Employee performance

All VIF values are below 10 (Kutner, et al, 2004) and all Tolerance values are above 0.1,
indicating that there is no significant multicollinearity among the independent variables.
Typically, a VIF value above 10 and a Tolerance value below 0.1 indicate multicollinearity,
which is not observed here.
Based on the VIF and Tolerance values obtained from the regression analysis, it can be
concluded that multicollinearity is not a problem for the variables training, Work-life balance,
Job satisfaction, and Organizational Commitment in predicting Employee Performance. This
means that these variables can be reliably used together in a regression model without
significant issues related to multicollinearity.
4.4 Demographic Statistics

The demographic statistics of the respondents are shown in this part depending on their

demographic traits. Gender, position, educational level, age, business type, and number of

employees of the respondents are among the demographic variables. The results are presented

on Table 4.4

Table 4.4: Demographic Characteristics of Respondents

Characteristics Frequency Percentage Cumulative Percentage
Gender
Female 72 32.9 32.9

Male 146 67.1 100

Total 218 100

Position Frequency Percentage Cumulative Percentage

Manager 81 37.2 37.2

Owner 137 62.8 100

Total 218 100

Age of Firm Frequency Percentage Cumulative Percentage

<5 198 90.8 90.8

6-10 20 9.20 100

11-20 0 0 0

>20 0 0 0
 Total 218 100

Educational Level Frequency Percentage Cumulative Percentage

First Degree 159 72.9 72.9

Post-Graduate 12 5.50 78.4

Others 47 21.60 100

Total 218 100

Type of Business Frequency Percentage Cumulative Percentage

Manufacturing 85 38.9 38.9

Services 133 61.1 100

Total 218 100

Number of Frequency Percentage Cumulative Percentage

Employees

10 – 49 181 83.1 83.1

50 – 199 37 16.9 100

Total 218 100

Above table reports the result of the demographic characteristics of the respondents of this

study. The data shows that the majority of the respondents are male (67.1%), while females

constitute 32.9% of the sample. This indicates a higher representation of male respondents in

the study. Among the respondents, 62.8% are owners, while 37.2% are managers. This

suggests that the sample includes more business owners than managers.

Most of the firms are relatively young, with 90.8% being less than 5 years old. A smaller

proportion (9.2%) of firms are between 6-10 years old, and none of the firms are older than
10 years. A majority of the respondents have a first degree (72.9%). Those with post-graduate

qualifications represent 5.5% of the sample, while 21.6% have other types of educational

backgrounds.

The respondents are mainly from the services sector (61.1%), with the remaining 38.9% from

the manufacturing sector. This indicates a higher prevalence of service-oriented businesses in

the sample. Most firms have between 10 and 49 employees (83.1%), whereas 16.9% of firms

have between 50 and 199 employees. This suggests that the sample primarily consists of

small-sized firms.

Variable Sample Mean Std. Deviation

Training 218 3.45 0.67

Workplace 218 3.78 0.74

Management

Organizational 218 3.65 0.71

Effective

Commitment

Employee 218 4.12 0.58

Performance

The table above reports the descriptive statistics of the variables of the study.

Training: The average training level provided is 3.45 with a standard deviation of 0.67,

indicating moderate training provision with some variability among respondents.

Workplace Management: The average effectiveness of workplace management is 3.78 with

a standard deviation of 0.74, suggesting a relatively high effectiveness with moderate
variability.
Organizational Effective Commitment: The average level of organizational commitment is
3.65 with a standard deviation of 0.71, indicating a relatively high commitment with
moderate variability.

Employee Performance: The average employee performance level is 4.12 with a standard
deviation of 0.58, indicating high performance with less variability among respondents.

These interpretations help understand the central tendencies and spread of the data for each
variable, which is crucial for further analysis and decision-making.
Measurement model
Note: Work place environment (WPM) and training (TRN) are the independent variables.
While organisational affective commitment (OAC) and employee performance (EP) are the
moderator and dependent variables respectively.

Construct Reliability and Validity

Construct Items Loadings AVE CR

Work place environment WPM1 0.766 0.600 0.947

WPM2 0.710
WPM3 0.817
WPM4 0.857
WPM5 0.724
WPM6 0.579
WPM7 0.806
WPM8 0.787
WPM9 0.837
WPM10 0.729
WPM11 0.804
WPM12 0.837
Training TRN1 0.771 0.613 0.888
TRN2 0.722
TRN3 0.833
TRN4 0.781
TRN5 0.804
Organisational affective commitment OAC1 0.672 0.552 0.895
OAC3 0.766
OAC4 0.816
OAC5 0.750
OAC6 0.626
OAC7 0.746
OAC8 0.805
Employee performance EP1 0.799 0.500 0.832
EP2 0. 744
EP3 0. 598
EP4 0.697
EP5 0.683

Note: Item OAC3 i.e. oorganisational affective commitment was deleted because it failed the
test.

Analysis:

Work Place environment:

Average Variance Extracted (AVE): 0.600

Composite Reliability (CR): 0.947

The AVE for Work Place environment is 0.600, which is above the acceptable threshold of
0.50, indicating that the construct explains more than half of the variance of its items. The CR
of 0.947 indicates excellent internal consistency and reliability.

Training:

Average Variance Extracted (AVE): 0.613

Composite Reliability (CR): 0.888

The AVE for Training is 0.613, indicating good convergent validity. The CR of 0.888 shows
that the construct has good internal consistency and reliability.

Organizational Affective Commitment:

Average Variance Extracted (AVE): 0.552

Composite Reliability (CR): 0.895

The AVE for Organizational Affective Commitment is 0.552, which meets the minimum
acceptable value, suggesting adequate convergent validity. The CR of 0.895 indicates good
reliability.

Employee Performance:

Average Variance Extracted (AVE): 0.500

Composite Reliability (CR): 0.832

The AVE for Employee Performance is exactly 0.500, which is the minimum acceptable
value, indicating that the construct explains half of the variance in its items. The CR of 0.832
suggests good reliability.

The constructs exhibit satisfactory reliability and validity, with AVE values indicating good
convergent validity (above 0.5) for all constructs. The CR values are all above 0.7,
demonstrating strong internal consistency. The constructs are, therefore, reliable and valid for
further analysis.

Literature Support:

This analysis aligns with the criteria established by Fornell and Larcker (1981), where:

AVE should be 0.5 or higher to indicate sufficient convergent validity. CR should be 0.7 or
higher to demonstrate good reliability of the constructs.
 Discriminant Validity using Fornell-larcker criterion
1 TRN 2 WPM 3 OAC 4 EP

1. Training 0.783

2. Work place environment 0.724 0.775

3. Organisational affective commitment 0.689 0.736 0.743

4. Employee performance 0.649 0.654 0.675 0.707

Analysis of Discriminant Validity

Training (TRN)

The square root of AVE (0.783) is greater than its correlations with other constructs (0.724

with WPM, 0.689 with OAC, 0.649 with EP). The Training construct has good discriminant

validity.

Work Place environment (WPM)

The square root of AVE (0.775) is greater than its correlations with other constructs (0.724

with TRN, 0.736 with OAC, 0.654 with EP). The Work Place environment construct has

good discriminant validity.

Organisational Affective Commitment (OAC)

The square root of AVE (0.743) is greater than its correlations with other constructs (0.689

with TRN, 0.736 with WPM, 0.675 with EP). The Organisational Affective Commitment

construct has good discriminant validity.

Employee Performance (EP)

The square root of AVE (0.707) is greater than its correlations with other constructs (0.649

with TRN, 0.654 with WPM, 0.675 with OAC). The Employee Performance construct has

good discriminant validity. The analysis using the Fornell-Larcker criterion demonstrates that

all constructs (Training, Work Place environment, Organisational Affective Commitment,

and Employee Performance) exhibit good discriminant validity. This indicates that each

construct is distinct and captures unique aspects of the variables being measured.

However, extant studies in the literature suggests that fornell and larcker criterion is

ineffective especially when the indicator loadings are between 0.65 and 0.85. Henselet et. Al,

(2015) recommended using the corelations, Heteriotrait-Monotrait ratio (HTMT) as a

substitute. Heir et. Al, (2021) defined the HTMT ratio as, “the mean of all correlations of

indicators across constructs measuring different constructs (i.e., the heteriotrait-heteromethod

correlations) relative to the mean of the average correlations of indicators measuring the same

construct (i.e the monotrait- heteromethod correlations).

Heir et. Al, (2021) and Henseller et. Al, (2015) proposed a threshold HTMT value of 0.90 for

structural models that are conceptually similar construct but a lower and more sophisticated

threshold HTMT value of 0.85 for structural models that have conceptually different

constructs.

: Heterotrait-Monotrait Ratio (HTMT)

Construct 1 2 3 4
1. Training
2. Work place environment 3. 0.724
Organisational affective management 0.689 0.736
4. Employee performance 0.649 0.654 0.675

Using these values, the HTMT values can be interpreted as follows:

HTMT (TRN, WPM) = 0.724
HTMT (TRN, OAC) = 0.689
HTMT (TRN, EP) = 0.649
HTMT (WPM, OAC) = 0.736
HTMT (WPM, EP) = 0.654
HTMT (OAC, EP) = 0.675
HTMT Interpretation
The threshold values for HTMT are typically below 0.9 (Henseler, Ringle, & Sarstedt,
2015).) indicates good discriminant validity. The constructs model demonstrate good
discriminant validity as all HTMT values are below the recommended threshold of 0.9. This
suggests that the constructs are distinct from each other, supporting the reliability of your
measurement model.

Structural Model

Bootstrapping is used to develop a model that best describes the population as a whole (Hair
et al, 2014). Bootstrapping was done on 205 cases utilizing 5000 subsamples. The structural
model of the independent variables effects on the dependent variables is shown

Note: Work place environment (WPM) and training (TRN) are the independent variables.
While organisational affective commitment (OAC) and employee performance (EP) are the
moderator and dependent variables respectively.
 Test of Hypotheses of the Study
The study tested for the effects of access to finance and leasing on performance. Thus, testing
the hypotheses of the study.
Direct Path Coefficient
Hypotheses Beta Value Standard Error T Stat P Value Decision
H01: WP-EP 0.571 0.099 5.767 0.000 Rejected

H02: TRN-EP 0.021 0.095 0.222 0.825 Accepted

R Square 62%

Analysis of Direct Path Coefficient Results

The direct path coefficient analysis provides insights into the relationships between the
constructs in the study. This analysis includes beta values, standard errors, t-statistics, and p-
values to determine the significance and strength of these relationships. Additionally, the R
Square value indicates the proportion of variance in the dependent variable (Employee
Performance) explained by the independent variables (Work place environment and
Training).
Hypothesis 1: Work place environment and Employee Performance (WP-EP)
Beta Value: 0.571
The beta value of 0.571 indicates a positive relationship between work place environment and
employee performance. Specifically, for each unit increase in work place environment,
employee performance increases by 0.571 units.
Standard Error: 0.099
The standard error is relatively small, suggesting that the estimate of the beta value is precise.
T Statistic: 5.767
A t-statistic of 5.767 is significantly greater than the critical value (typically around 1.96 for a
95% confidence level), indicating that the relationship is statistically significant.
P Value: 0.000
The p-value of 0.000 is less than 0.05, indicating that the null hypothesis (no relationship) can
be rejected. Therefore, there is a statistically significant positive effect of work place
environment on employee performance.
Decision: The hypothesis H01 is supported. There is a significant positive relationship
between work place environment and employee performance.

Hypothesis 2: Training and Employee Performance (TRN-EP)

Beta Value: 0.021
The beta value of 0.021 indicates a very weak positive relationship between training and
employee performance. This suggests that for each unit increase in training, employee
performance increases by only 0.021 units.
Standard Error: 0.095
The standard error is relatively small, suggesting that the estimate of the beta value is precise.
T Statistic: 0.222
A t-statistic of 0.222 is significantly lower than the critical value (typically around 1.96 for a
95% confidence level), indicating that the relationship is not statistically significant.
P Value: 0.825
The p-value of 0.825 is greater than 0.05, indicating that the null hypothesis (no relationship)
cannot be rejected. Therefore, there is no statistically significant effect of training on
employee performance.
Decision: The hypothesis H02 is not supported. Training does not have a significant impact
on employee performance.
R Square: 62%
The R Square value of 62% indicates that 62% of the variance in employee performance can
be explained by the independent variables (work place environment and training). This is a
relatively high R Square value, suggesting that the model has good explanatory power.

Interpretation
H01 (WP-EP): The relationship between work place environment and employee performance
is strong and statistically significant. This indicates that improvements in work place
environment are likely to lead to better employee performance.
H02 (TRN-EP): The relationship between training and employee performance is very weak
and not statistically significant. This suggests that training, as measured in this study, does not
have a noticeable impact on employee performance. Further investigation might be needed to
explore other aspects of training or different variables that could influence this relationship.
R Square: The model explains 62% of the variance in employee performance, which is
substantial and indicates that the independent variables included in the model are good
predictors of employee performance.

Effect Size of Exogenous Variables

Table 4.10: Effect Size for Direct Relationships
Construct f2 Effect Size

Work place environment 0.19 Medium

Training 0.06 Small

Effect Size of Exogenous Variables

The effect size (f²) is a measure of the strength of the relationship between exogenous
(independent) variables and the endogenous (dependent) variable. In this context, we are
evaluating the impact of 'Work Place Environment' and 'Training' on the dependent variable,
which we can infer as 'Employee Performance' based on previous analyses.
Interpretation of Effect Size (f²)
According to Cohen's guidelines, the interpretation of f² values is as follows:
Small effect: 0.02
Medium effect: 0.15
Large effect: 0.35
Analysis
Work Place Environment
f² = 0.09: The effect size of 0.09 indicates a small effect of the work place environment on the
dependent variable. This suggests that changes in the work place environment do contribute
to employee performance, but not to a large extent. Although small, the effect is meaningful
and should be considered in the broader context of workplace improvements.
Training
f² = 0.00: The effect size of 0.00 indicates no effect of training on the dependent variable.
This finding suggests that, within this sample and context, training does not significantly
contribute to employee performance. This result aligns with the hypothesis test (H02: TRN-
EP) where the p-value was not significant (0.825).
Both constructs, 'Work Place Environment' and 'Training', show different levels of effect size.
The 'Work Place Environment' has a small effect size, indicating it does play a role in
employee performance. However, 'Training' does not show any effect size, implying it has no
significant impact on employee performance in this study.
Work Place Environment: Given its small but significant effect, improvements in the work
place environment can positively impact employee performance. Organizations should
continue to invest in enhancing workplace conditions.
Training: The lack of effect size suggests that the current training programs might not be
effective in improving employee performance. Organizations might need to reassess and
redesign their training programs or explore alternative methods to improve employee
performance.
The analysis of the effect sizes underscores the importance of the work place environment in
influencing employee performance, albeit to a small extent. Conversely, the current training
initiatives do not appear to be effective in this regard. A comprehensive approach that
includes improving the work place environment while reevaluating training programs could
be more beneficial for enhancing employee performance.

Predictive Relevance of Exogenous Variables

The study utilised the Stone-Geisser’s Q 2 value to evaluate the predictive relevance of the

exogenous variables.

Predictive Relevance of Exogenous Variables

Construct SSO SSE Q2 = 1-SSE/SSO

Employee Performance 0.877 0.352 0.278

The table presents the result of cross-validated redundancy of the relationships between
access to finance, leasing and SME performance. Q2 is greater than zero which shows the
predictive relevance of the path model. The model has high degree of predictive relevance on
performance (Cohen, 1988).

Test of Hypotheses of Moderating Relationships

The study examines the moderating role of organizational affective commitment in the

relationship between workplace environment and employee performance.

Structural Model: Bootstrapping Results for Moderating Effect

Hypotheses Beta S. Error T-Stat P-Value Decision

H03 OAC x WPM->EP 0.395 0.046 8.749 0.000 Rejected

H04 OAC x TRN->EP 0.154 0.076 2.042 0.080 Accepted

The table presented outlines the bootstrapping results for testing the moderating effects in a
structural model. Specifically, it examines the interaction effects of Organizational affective
commitment (OAC) with Work Performance Measurement (WPM) and Training (TRN) on
Employee Performance (EP).
H03 (OAC x WPM -> EP): The interaction between Organizational affective commitment and
Work Performance Measurement has a Beta coefficient of 0.395 with a standard error of
0.154. The T-statistic of 8.749 indicates that this effect is statistically significant at the 0.05
level (p = 0.000), leading to the decision to reject the null hypothesis (Decision: 0.000).
H04 (OAC x TRN -> EP): The interaction between Organizational affective commitment and
Training has a Beta coefficient of 0.046 with a standard error of 0.076. The T-statistic of
2.042 shows that this effect is not statistically significant at the 0.05 level (p = 0.080), leading
to the decision to accept the null hypothesis (Decision: 0.080).

Assessment of the Effect Size of the interaction terms: F-Square

Construct f2(PRF) Effect Size

OAC x WPM->EP 0.40 Large

OAC x TRN->EP 0.12 Small

The interaction between OAC and WPM (OAC x WPM -> EP) shows a relatively strong
effect size (f square = 0.40), indicating that this interaction significantly contributes to
explaining the variance in EP beyond the main effects. In contrast, the interaction between
OAC and TRN (OAC x TRN -> EP) has a smaller effect size f square = 0.12), suggesting that
while it is statistically significant, its contribution to explaining variance in EP is
comparatively weaker.
Construct cross validated Redundancy
Construct Q² (=1-SSE/SSO)

Employes Performance 0.271

The study utilised the Stone-Geisser’s Q2 value to evaluate the predictive relevance of the
exogenous variables. This Q² value of 0.271 indicates the proportion of variance in employee
performance that is predicted by the model beyond what would be predicted by chance. Q2
=0.271 suggests a moderate level of explanatory power, indicating that the model reasonably
predicts Employee Performance based on the included constructs.

Summary of Findings: Hypotheses Testing Results

Hypotheses Findings

1. Work place environment->Employee performance Rejected

2. Training-> Employee Performance

3. Organisational affective commitment x Work place environment -> Employee Accepted

Performance Rejected

4. Organisational affective commitment x Training -> Employee Performance

Accepted

Discussions of Findings
The study aimed to investigate various hypotheses related to factors influencing Employee
Performance (EP), focusing on Workplace Environment (H1), Training (H2), and their
interactions with Organizational Affective Commitment (OAC) (H3 and H4). This discussion
analyses the findings derived from hypothesis testing.
Hypothesis 1: Workplace Environment -> Employee Performance
Finding: Rejected
The hypothesis testing indicates that Workplace Environment significantly influences
Employee Performance. This finding aligns with existing literature that emphasizes the
impact of a supportive and conducive workplace environment on employee outcomes. A
positive workplace environment, characterized by factors such as organizational culture,
physical workspace, and interpersonal relationships, tends to enhance employee motivation,
job satisfaction, and overall performance.
Hypothesis 2: Training -> Employee Performance
Finding: Accepted
Contrary to expectations, the hypothesis testing does not support a significant positive
relationship between Training (TRN) and Employee Performance (EP). Training programs
are typically designed to equip employees with the necessary skills, knowledge, and
competencies to perform their jobs effectively. However, the acceptance of this hypothesis
suggests that in this study, training initiatives did not demonstrate a statistically significant
impact on improving employee performance outcomes. The findings indicate that while
training may enhance individual employee capabilities, it did not show a significant
contribution to organizational effectiveness by aligning employee skills with job
requirements and strategic objectives. This implies that organizations investing in
comprehensive training programs may not necessarily experience improved performance
metrics, employee satisfaction, and retention rates as hypothesized.
Hypothesis 3: Organisational Affective Commitment x Workplace Environment -> Employee
Performance
Finding: Rejected
Consistent with expectations, the hypothesis testing results indicate a significant positive
interaction between Organizational Affective Commitment (OAC) and Workplace
Environment (WPE) in predicting Employee Performance (EP). Organizational Affective
Commitment refers to employees' emotional attachment, loyalty, and identification with their
organization, factors that are expected to synergistically interact with a conducive workplace
environment to enhance performance. The rejection of this null hypothesis suggests that the
combined effect of OAC and WPE provides additional explanatory power beyond their
individual effects on employee performance. This finding underscores the importance of
fostering a supportive workplace environment that nurtures employees' emotional
commitment and attachment to the organization, thereby enhancing overall performance
outcomes. Future research may further explore specific dynamics and contexts where this
interaction is particularly influential, providing insights into optimizing organizational
strategies to promote employee engagement and productivity.
Hypothesis 4: Organisational Affective Commitment x Training -> Employee Performance
Finding: Accepted
In contrast to Hypothesis 3, the hypothesis testing results do not support a significant
prediction of Employee Performance (EP) by the interaction between Organizational
Affective Commitment (OAC) and Training (TRN). Organizational Affective Commitment
refers to employees' emotional attachment, loyalty, and identification with their organization,
which was expected to interact positively with training initiatives to enhance performance.
However, the acceptance of this null hypothesis suggests that in this study, the combined
effect of OAC and TRN did not demonstrate a statistically significant impact on improving
employee performance outcomes. This finding indicates that while organizational
commitment and training individually impact employee performance, their interaction did not
provide additional explanatory power beyond their main effects. Further exploration into
specific conditions or contexts where OAC and TRN interact differently than hypothesized
may provide insights into optimizing training strategies to maximize their impact on
employee engagement and performance.
and growth through training programs.

CHAPTER FIVE
SUMMARY, CONCLUSION AND RECOMMENDATIONS
Summary

The study investigated the influence of Workplace Environment (H1), Training (H2), and
their interactions with Organizational Affective Commitment (OAC) (H3 and H4) on
Employee Performance (EP). Findings revealed nuanced relationships. Workplace
Environment significantly impacts EP, aligning with literature emphasizing its role in
enhancing employee motivation and job satisfaction. However, Training did not show a
significant positive relationship with EP, suggesting that while it enhances individual
capabilities, its impact on organizational effectiveness was inconclusive in this study. The
interaction between OAC and Workplace Environment (H3) positively predicted EP,
highlighting the synergistic effects of emotional attachment and a supportive workplace.
Conversely, the interaction between OAC and Training (H4) did not significantly predict EP,
indicating that their combined effect did not exceed their individual impacts. These findings
underscore the complexity of factors influencing EP and suggest avenues for optimizing
organizational strategies to foster employee engagement and productivity.

Conclusion
The findings from the hypotheses testing provide valuable insights into the factors
influencing Employee Performance within the organizational context studied. Acceptance and
rejection of hypotheses indicate the varying impact of different organizational factors on
employee outcomes.

Implications of Findings
The rejection of null Hypothesis 1 and acceptance of Hypothesis 2 suggest important
implications for organizational strategies. Firstly, the significant influence of Workplace
Environment (H1) underscores the critical need for organizations to invest in cultivating
supportive and conducive work environments. Such environments, characterized by positive
organizational culture, well-designed physical spaces, and strong interpersonal relationships,
can significantly enhance employee motivation, job satisfaction, and overall performance.
Organizations should prioritize initiatives that create and maintain these supportive
environments to maximize employee well-being and productivity.
Secondly, while Training (H2) did not show a significant direct relationship with Employee
Performance (EP) in this study, it remains crucial for organizations to carefully design and
implement training programs. Despite the lack of statistical significance in this context,
effective training remains pivotal in equipping employees with essential skills, knowledge,
and competencies. Therefore, organizations should continue to invest in comprehensive
training initiatives that align closely with organizational goals and individual employee
development needs. Continuous evaluation and adaptation of training strategies are essential
to ensure they effectively contribute to enhancing employee capabilities and organizational
effectiveness over time.
The rejection of null Hypothesis 3 suggests that Organizational Affective Commitment
(OAC) and Workplace Environment (WPE) significantly interact in predicting Employee
Performance (EP). This significant interaction underscores the need for further exploration
into how these factors work together across different organizational settings. While this study
has established their combined effect, future research could delve deeper into specific
contexts or conditions where this interaction might manifest differently. Understanding these
dynamics can provide valuable insights into optimizing organizational strategies to foster a
supportive workplace environment that enhances employee commitment and performance.
On the other hand, the acceptance of null Hypothesis 4 highlights the potential benefits of
integrating organizational commitment with effective training strategies to optimize
employee performance and organizational success. Although the study did not find a
significant combined effect of OAC and Training (TRN) on EP, organizations can still
leverage both factors individually to enhance employee engagement and job satisfaction.
Integrating organizational commitment initiatives with tailored training programs that address
specific skill gaps and developmental needs can effectively contribute to improving employee
capabilities and aligning them with organizational objectives. These implications underscore
the importance of a nuanced approach in organizational management, where understanding
the interplay of factors influencing employee performance can guide strategic decisions
aimed at maximizing organizational effectiveness and fostering a positive work environment.
This study contributes to the existing literature by providing empirical evidence on the
relationships between organizational factors and employee performance, particularly
emphasizing the interactive effects of organizational commitment and training.
Limitations
It's important to acknowledge potential limitations such as sample size, specific contextual
factors, and measurement scales used, which may affect the generalizability of findings to
broader organizational contexts.
Recommendations
Based on the findings, several recommendations can be made for practice and further
research:
Enhancing Workplace Environment
Organizations should prioritize creating and maintaining a positive workplace environment
characterized by supportive leadership, clear communication, and opportunities for growth
and development. This can contribute to higher employee motivation, satisfaction, and
productivity.
Investing in Training and Development
Continuous investment in training and development programs tailored to organizational needs
and employee growth areas is crucial. Training should not only focus on technical skills but
also on soft skills and leadership development to enhance overall performance outcomes.
Integrating Organizational Commitment
Strategies to foster and sustain organizational commitment among employees should be
integrated into HR policies and practices. This can include recognition programs, career
development opportunities, and transparent communication to strengthen employee loyalty
and engagement.
Further Research Directions
Future research could explore additional moderators or mediators that may influence the
relationships identified in this study. For instance, the role of leadership style, organizational
culture, or industry-specific factors in shaping employee performance outcomes.
Longitudinal studies could provide insights into the sustainability and long-term impact of
organizational interventions on employee performance metrics.
The findings from this study provide valuable implications for organizations aiming to
enhance employee performance through strategic investments in workplace environment,
training programs, and organizational commitment. By understanding the nuanced
relationships between these factors, organizations can tailor their policies and practices to
create a conducive work environment that promotes employee engagement, satisfaction, and
productivity. Through continuous evaluation and adaptation of organizational strategies based
on empirical evidence, organizations can position themselves for sustained success in a
competitive marketplace while fostering a positive work culture that supports employee well-
being and professional growth. This comprehensive discussion integrates the findings,
implications, limitations, and recommendations based on the hypotheses testing results,
offering a robust framework for understanding and optimizing employee performance within
organizational contexts.

Variables Number of missing values

Employee Performance 2
Training 1
Organisational 2
effective commitment
2
Work place
environment 7 (out of 9,967 data points)

Total 0.07%

Percentage

Construct Cross-loading
EP OAC TRN WPM
 EP1 0.799 0.639 0.674 0.677
EP2 0.744 0.468 0.409 0.535
EP3 0.598 0.410 0.375 0.421
EP4 0.697 0.473 0.483 0.554
EP5 0.683 0.472 0.441 0.494
OAC1 0.533 0.672 0.444 0.506
OAC3 0.578 0.766 0.589 0.606
OAC4 0.573 0.816 0.527 0.585
OAC5 0.513 0.750 0.511 0.597
OAC6 0.367 0.626 0.414 0.435
OAC7 0.528 0.746 0.615 0.627
OAC8 0.542 0.805 0.551 0.611
TRN1 0.585 0.564 0.771 0.629
TRN2 0.418 0.491 0.722 0.548
TRN3 0.562 0.547 0.833 0.766
TRN4 0.516 0.576 0.781 0.710
TRN5 0.589 0.579 0.804 0.817
WPM1 0.562 0.547 0.833 0.766
WPM10 0.544 0.573 0.580 0.729
WPM11 0.578 0.566 0.783 0.804
WPM12 0.613 0.634 0.654 0.837
WPM2 0.516 0.576 0.781 0.710
WPM3 0.589 0.579 0.804 0.817
WPM4 0.638 0.635 0.687 0.857
WPM5 0.552 0.591 0.589 0.724
WPM6 0.562 0.498 0.495 0.579
WPM7 0.664 0.615 0.744 0.806
WPM8 0.677 0.675 0.694 0.787
WPM9 0.610 0.618 0.656 0.837

Reference

Kutner MH, Nachtsheim CJ, Neter J, Li W, (2004). Applied Linear Statistical Models, 5th
edition, pp. 15, McGraw‐Hill/Irwin.