CHAPTER 4
DATA ANALYSIS AND PRESENTATION
4.1 INTRODUCTION
4.2 RESPONSE RATE
The study targeted a total of 8 hospital firms, out of which 6 successfully participated, yielding a
response rate of 75%. According to Mugenda and Mugenda (2003), a response rate of 70% and
above is considered very good, indicating strong participation. Similarly, Sataloff and Vontela
(2021) classify response rates above 70% as excellent, further affirming the adequacy and
reliability of the data collected in this study. Therefore, the 75% response rate attained is both
commendable and sufficient to support the study’s findings.
4.3 Descriptive Statistics
This section presents central tendency measures (Mean, Median, and Mode) for the dependent
and independent variables.
4.3.1 Mean
Table 4.2 Mean
Variable Operating Credit Risk Regulatory Operational
Margin (%) Score Risk Score Risk Score
Mean -21.46 3.38 3.17 3.96
The mean of -21.46% shows that average financial performance is poor, indicating widespread
losses.
The mean of 3.38 average credit risk is moderate.
The mean of Regulatory risk is 3.17 which slightly lower than credit risk, but still moderate
4.3.2 Median
Table 4.2 Median
Variable Operating Credit Risk Regulatory Operational
Margin (%) Score Risk Score Risk Score
Median -11.82 3.63 3.63 4.13
Half of hospitals perform better than -11.82%, but half are worse. The median being less
negative than the mean suggests a few extreme losses skew the data.
Half of hospitals have scores above 3.63, indicating moderate-to-high risk is typical.
Median of 3.63 Confirms that most hospitals cluster toward higher risk.
4.3.3 Mode
Variable Operating Credit Risk Regulatory Operational
Margin (%) Score Risk Score Risk Score
Mode 0.00 1.00 1.00 2.50
The mode of 0.00% shows that the most common result is break-even, implying some hospitals
avoid losses despite sector-wide challenges.
The mode of 1.00 suggests a subgroup with minimal credit risk.
The mode of 1.00 shows that a low-risk subgroup exists.
4.4 Inferential Statistics
The section below combines the model summary, ANOVA, and coefficients for each risk
exposure into one comprehensive table per variable.
4.4.1 Regression Results: Financial Performance vs. Credit Risk
Model R / R² / Adj R² Std. Error / F B B (Credit Interpretation
/ Sig. (Constant) Risk) /
/ t / Sig. Beta / t /
Sig.
Model R=0.629, Std. - - Explains
Summary R²=0.396, Adj. Error=23.3673 model fit and
R²=0.094 variation
explained.
ANOVA SS F=1.312, - - Tests overall
Reg=716.355, Sig.=0.371 significance of
SS the regression.
Resid=1092.061
Coefficients - - -132.637 / - 24.724 / Coefficient
1.499 / 0.629 / impact of
0.273 1.145 / Credit Risk on
0.371 financial
performance.
A simple linear regression was conducted to assess the effect of credit risk on the financial
performance of hospitals in Nairobi. The analysis yielded an R value of 0.629, indicating a
moderate positive correlation between credit risk scores and operating margin. This suggests that
higher credit risk scores tend to be associated with higher financial performance.
The R Square value was 0.396, showing that 39.6% of the variation in financial performance can
be explained by credit risk. However, the Adjusted R Square was 0.094.
The ANOVA results for the model examining credit risk and financial performance yielded an
F-statistic of 1.312 and a p-value (Sig.) of 0.371. Since the p-value is greater than 0.05, the
regression model is not statistically significant. This indicates that credit risk is not a
significant predictor of financial performance among the hospitals studied. The constant
(intercept) value B = -132.637 indicates that when the credit risk score is zero, the predicted
operating margin is -132.64%. This suggests hospitals with no credit risk would still be
expected to operate at a significant loss. The coefficient for Credit Risk Score is B = 24.724,
meaning that for every one-unit increase in the credit risk score, the operating margin
increases by approximately 24.72%. However, the t-value = 1.145 and p-value = 0.371
indicate that this effect is not statistically significant at the 0.05 level. Thus, we cannot
confidently state that credit risk significantly affects financial performance.
4.4.2 Regression Results: Financial Performance vs. Regulatory Risk
Model R / R² / Adj R² Std. Error / F / B B Interpretation
Sig. (Constant) / (Regulatory
t / Sig. Risk) /
Beta / t /
Sig.
Model R=0.548, Std. - - Explains
Summary R²=0.3, Adj. Error=25.15724 model fit and
R²=-0.05 variation
explained.
ANOVA SS F=0.857, - - Tests overall
Reg=542.643, Sig.=0.452 significance
SS of the
Resid=1265.773 regression.
Coefficients - - -74.713 / - 12.369 / Coefficient
1.569 / 0.548 / impact of
0.257 0.926 / Regulatory
0.452 Risk on
financial
performance.
Another regression analysis was performed to investigate the relationship between regulatory
risk and financial performance. The regression produced an R value of 0.548, suggesting a
moderate positive correlation between the two variables.
The R Square was 0.300, meaning that 30% of the variation in financial performance can be
explained by differences in regulatory risk. However, the Adjusted R Square was -0.050.
For the regression model assessing regulatory risk and financial performance, the F-statistic was
0.857 and the p-value was 0.452. Again, since the p-value exceeds the 0.05 threshold, the model
is not statistically significant. Therefore, we cannot conclude that regulatory risk has a
significant impact on hospital financial performance in this sample. The constant is B = -
74.713, suggesting that in the absence of regulatory risk, the expected operating margin would be
-74.71%, indicating poor financial performance by default. The coefficient for Regulatory Risk
is B = 12.369, meaning that each one-unit increase in regulatory risk is associated with an
increase in operating margin of 12.37%, holding all else constant. However, the associated p-
value is 0.452, showing that this effect is not statistically significant. The predictor does not
have a meaningful influence on financial performance based on this sample.
4.4.3 Regression Results: Financial Performance vs. Operational Risk
Model R / R² / Adj R² Std. Error / F / B B Interpretation
Sig. (Constant) / (Operational
t / Sig. Risk) /
Beta / t /
Sig.
Model R=0.354, Std. - - Explains
Summary R²=0.126, Adj. Error=28.1190 model fit and
R²=-0.312 7 variation
explained.
ANOVA SS F=0.287, - - Tests overall
Reg=227.052, Sig.=0.646 significance
SS of the
Resid=1581.364 regression.
Coefficients - - -113.2 / - 19.06 / Coefficient
0.746 / 0.354 / impact of
0.534 0.536 / Operational
0.646 Risk on
financial
performance.
A final regression analysis explored the impact of operational risk on the financial performance
of hospitals. The model returned an R value of 0.354, representing a weak positive correlation.
The R Square was 0.126, indicating that only 12.6% of the variance in financial performance can
be attributed to operational risk. Additionally, the Adjusted R Square was -0.312.
The ANOVA for the model involving operational risk produced an F-statistic of 0.287 and a p-
value of 0.646. This result is clearly not statistically significant (p > 0.05), which implies that
operational risk does not significantly predict financial performance within the context of the
hospitals included in the study.
The constant in this model is B = -113.200, implying that if operational risk were zero, hospitals
would still be predicted to experience a negative operating margin of -113.20%. The coefficient
for Operational Risk is B = 19.060, suggesting that each one-unit increase in operational risk
leads to a 19.06% increase in operating margin, all else being equal. Yet, the p-value of 0.646
shows that this effect is not statistically significant, meaning the relationship could be due to
chance and not a true predictive influence.