# Interpreting Results Problem

#### eli

##### New Member
Hello there,
I encountered difficulties during interpretation of the results, relationship is positive but insignificant. for DER and ICR variable.
i would appreciate a help.
Thanks

#### Karabiner

##### TS Contributor
DER and ICR show statistically significant relationships with the dependent variabe, while DR does not. What is your specific problem with that?

With kind regards

Karabiner

#### eli

##### New Member
DER and ICR show statistically significant relationships with the dependent variabe, while DR does not. What is your specific problem with that?

With kind regards

Karabiner
I dont kow how to interpret that, for example > DER show positive relation with DR but its significance is low. For conclusion this Variale is or is not an important determinant of financial performance of quoted companies. So i come in conclusion for Hypotheses.
Best regards,
Eli

#### Dason

##### Ambassador to the humans
I think you are misunderstanding. The Sig. Column is almost certainly showing the p-value and for p-values lower gets interpreted as "more significant". Please read up on p-values and how to interpret them.

#### hlsmith

##### Less is more. Stay pure. Stay poor.
Also, you keep saying things are positive. News flash, the coefficients are always going to be positive or negative. The next thing that comes into play is that given your data and model, can you conclusively say that the positive or negative value may not be closure or congruent to zero given repeated sampling of the super population? This is what your pvalue is getting at, the probability of having an effect that large or larger if the null was true and conditional on the model and data realization. So a positive or negative coefficient coupled with a small p-value provide information in regards to rejecting the null of no effect.

#### eli

##### New Member
Also, you keep saying things are positive. News flash, the coefficients are always going to be positive or negative. The next thing that comes into play is that given your data and model, can you conclusively say that the positive or negative value may not be closure or congruent to zero given repeated sampling of the super population? This is what your pvalue is getting at, the probability of having an effect that large or larger if the null was true and conditional on the model and data realization. So a positive or negative coefficient coupled with a small p-value provide information in regards to rejecting the null of no effect.
It is little confusing for me because its my first time interpreting spss results, i would be very thankfull if you let me a mail so i sent you all the documet to help me out.
With kind regards,
Eli

#### eli

##### New Member
DER and ICR show statistically significant relationships with the dependent variabe, while DR does not. What is your specific problem with that?

With kind regards

Karabiner
i would appreciate if you help me out with interpreting some results. i can sent in email the document.
With kind regards

#### hlsmith

##### Less is more. Stay pure. Stay poor.
@eli - I am happy to help out where I can. Though, you can feel free to just post subsequent inquiry on this site or if for some reason you have a question you are unable to publicly share you can send a personal message through this site, probably no attachments on a PM allowed.

Yes, I can understand how things can be tricky when reviewing output for the first time!

#### eli

##### New Member
Objectives of the study
The main objective of the study is to investigate the effect offinancial leverage on financial performance of companies with particular reference to quoted companies in Country. The specific objectives of this study are:
 To examine the effect of debt ratio (DR) on Return on Assets(ROA) of quoted companies in the country.
 To determine whether debt- equity ratio (DER) have any effect on Return on Assets (ROA) of quoted companies .
 To establish if there is any effect of interest coverage ratio (ICR) on Return on Assets (ROA) of quoted companies in coutry.

Statement of Hypotheses Based on the objectives of the study, the following hypotheses were developed:
H1: There is a significant effect of Debt Ratio (DR) onReturn on Assets (ROA) of quoted companies in the country.
H2: Debt to Equity Ratio (DER) has a significant effect on Return on Assets (ROA) of quoted companies in country.
H3: There is a significant effect of interest coverage ratio (ICR) on Return on Assets (ROA) of quoted companies in country.

Methodology
The research work focuses on the empirical analysis of the effect of financial leverage on financial performance in some selected companies in country. The ex-post factor research design was used. . This research relied heavily on historic data as data used in the analysis were generated from annual financial reports of the selected companies , a period of 3 years. The variables that were tested in this study were Return on Assets (ROA), Debt ratio (DR), Debt-equity-ratio (DER) and Interest coverage ratio (ICR).In this study, Financial Performance proxy Return on Assets (ROA) is our dependent variable while Financial Leverage measured by DR, DER and ICR are our independent variables.

The population of this research work was five (5) food companies in country.

Method of data Analysis
Descriptive analysis was firstly applied todescribe relevant aspects of financial leverage and provided detailed information about each relevant variable. Correlation models, specifically Pearson correlation were applied to measure the degree of association between different variables under consideration while regression analysis was applied to examine the relationship of independent variables with dependent variable and to know the effect of selected independent variableson financial performance. By using this method, researchers will be able to identify the significant of each explanatory variable to the model and also the significance of the overall model.The model used was multiple regressions (more than one independent variables). The researcher also used Ordinary Least Squares (OLS) method for analysis of hypotheses stated in a multiple form. For this purpose of analysis the MS Excel Software was used to analyse financial data and SSPS Software used to run the regression.

I would appreciate if you help me out interpreting this tables! thankss

#### eli

##### New Member
@eli - I am happy to help out where I can. Though, you can feel free to just post subsequent inquiry on this site or if for some reason you have a question you are unable to publicly share you can send a personal message through this site, probably no attachments on a PM allowed.

Yes, I can understand how things can be tricky when reviewing output for the first time!
Objectives of the study
The main objective of the study is to investigate the effect offinancial leverage on financial performance of companies with particular reference to quoted companies in Country. The specific objectives of this study are:
 To examine the effect of debt ratio (DR) on Return on Assets(ROA) of quoted companies in the country.
 To determine whether debt- equity ratio (DER) have any effect on Return on Assets (ROA) of quoted companies .
 To establish if there is any effect of interest coverage ratio (ICR) on Return on Assets (ROA) of quoted companies in coutry.

Statement of Hypotheses Based on the objectives of the study, the following hypotheses were developed:
H1: There is a significant effect of Debt Ratio (DR) onReturn on Assets (ROA) of quoted companies in the country.
H2: Debt to Equity Ratio (DER) has a significant effect on Return on Assets (ROA) of quoted companies in country.
H3: There is a significant effect of interest coverage ratio (ICR) on Return on Assets (ROA) of quoted companies in country.

Methodology
The research work focuses on the empirical analysis of the effect of financial leverage on financial performance in some selected companies in country. The ex-post factor research design was used. . This research relied heavily on historic data as data used in the analysis were generated from annual financial reports of the selected companies , a period of 3 years. The variables that were tested in this study were Return on Assets (ROA), Debt ratio (DR), Debt-equity-ratio (DER) and Interest coverage ratio (ICR).In this study, Financial Performance proxy Return on Assets (ROA) is our dependent variable while Financial Leverage measured by DR, DER and ICR are our independent variables.

The population of this research work was five (5) food companies in country.

Method of data Analysis
Descriptive analysis was firstly applied todescribe relevant aspects of financial leverage and provided detailed information about each relevant variable. Correlation models, specifically Pearson correlation were applied to measure the degree of association between different variables under consideration while regression analysis was applied to examine the relationship of independent variables with dependent variable and to know the effect of selected independent variableson financial performance. By using this method, researchers will be able to identify the significant of each explanatory variable to the model and also the significance of the overall model.The model used was multiple regressions (more than one independent variables). The researcher also used Ordinary Least Squares (OLS) method for analysis of hypotheses stated in a multiple form. For this purpose of analysis the MS Excel Software was used to analyse financial data and SSPS Software used to run the regression.

I would appreciate if you help me out interpreting this tables! thankss

#### hlsmith

##### Less is more. Stay pure. Stay poor.
I will look at this tomorrow,but it you have multiple va riable in the model your hypotheses should have addition text saying, while controlling for xyz.

#### eli

##### New Member
I will look at this tomorrow,but it you have multiple va riable in the model your hypotheses should have addition text saying, while controlling for xyz.
thank you

#### hlsmith

##### Less is more. Stay pure. Stay poor.
If there are 5 food companies, why is the n=15?

What are your specific questions in regards to the output. It looks straightforward. Descriptive stats, correlations, regression results.

#### eli

##### New Member
If there are 5 food companies, why is the n=15?

What are your specific questions in regards to the output. It looks straightforward. Descriptive stats, correlations, regression results.
5 companies during 3 years.
I would like to help me interpret the tables

#### hlsmith

##### Less is more. Stay pure. Stay poor.
Follow-up, who did these analyses? 5 companies over three years, would mean you should control for company or company contributing multiple observations if the OLS included the 15 values. Potential issue if this didn't occur!!

What don't you understand about the tables You know what mean, std, n's are right? You know what correlations are right? Is it just the regression output, because there are lots of online resources saying how to interpret SPSS regression output. You should review those and then ask us direct questions about whether your interpretations are correct or not.

#### eli

##### New Member
Follow-up, who did these analyses? 5 companies over three years, would mean you should control for company or company contributing multiple observations if the OLS included the 15 values. Potential issue if this didn't occur!!

What don't you understand about the tables You know what mean, std, n's are right? You know what correlations are right? Is it just the regression output, because there are lots of online resources saying how to interpret SPSS regression output. You should review those and then ask us direct questions about whether your interpretations are correct or not.
TABLE 4

The calculated ratio of the debt ratio (DR) shows that DR has good and positive return on assets (ROA). The significant positive report indicates that the debt ratio (DR) of domestic quoted companies can positively influence the financial performance of the pharmaceutical industry. However, its significance level of 0.527 indicates that tc (DR) is statistically insignificant.
Thus, the weight of evidence suggests that we accept H1 because there is an important effect of the debt-repayment (RO) ratio of the companies quoted. This means that a change in debt ratio virtually affects the financial performance of companies.
Moreover, it shows that tc (DER) stands at 3.496> t * 2 confirming that it is statistically important for the financial performance of the quoted companies. This indicator indicates that the debt-to-equity ratio (DER) has a positive relationship and statistically points to the financial performance of the industry significantly. However, its level of importance at 0.005 makes (DER) statistically insignificant. The weight of evidence, therefore, suggests Hi to be accepted. This means that the debt-to-equity ratio (DER) has an effect on the Return on Assets (ROA) of the quoted companies.
Finally, the coefficient outlined above reveals that the interest coverage ratio (ICR) has positive relationships and statistically affects the financial performance of companies. Given that the t-account of 4.761> t * 2, we confirm the statistically significant effect of the interest coverage ratio (ICR). This confirmation is strengthened with the value p- 0.001 <0.05 the value level of significance. Thus, the weight of evidence suggests that the zero hypothesis (Ho) is rejected and the alternative hypothesis (Hi) is accepted. This implies that it has a significant effect on the ICR on Return on Assets (ROA) of domestic quoted companies. So, companies use interest coverage to finance their company's expansion. So the test results described below provide significant assurance for the results and the multiple regression equation is as: ROA = -0.043 +0.028 (DR) +0.008 (DER) + 0.012 (ICR) + ɛi

#### hlsmith

##### Less is more. Stay pure. Stay poor.
@eli - your interpretations seem reasonable enough. B = beta or the relationship. So DR has the largest association, but given the large SE value you can't rule out that the association may equal "0" -> thus the high pvalue (AKA "Sig").

#### eli

##### New Member
@eli - your interpretations seem reasonable enough. B = beta or the relationship. So DR has the largest association, but given the large SE value you can't rule out that the association may equal "0" -> thus the high pvalue (AKA "Sig").
so are the interpretations alright?

#### hlsmith

##### Less is more. Stay pure. Stay poor.
Well without staring real hard they seemed correct. It is easy to get lost given your lack of spacing in the that big glob of text. Interpretations are pretty straightforward. If small p-value there is reasonable doubt in the null hypothesis of no association. If the coefficient is negative there is a negative association and vice versa. You just need to write out the results like:

A 1 unit increase in DER is associated with a 0.008 increase in ROA. You also need to think about whether a 0.008 increase is contextually important. So if you are on diet A and your average weight loss was 0.05 kg and it was statistically significant who really cares, since A cost money and effort.

Lastly, are you still allowing each group to contribute more than one value. An assumption of linear regression (OLS) is that the observations are independent. If you have two time dependent values from the same group you are breaking a key assumption of OLS.

#### eli

##### New Member
Well without staring real hard they seemed correct. It is easy to get lost given your lack of spacing in the that big glob of text. Interpretations are pretty straightforward. If small p-value there is reasonable doubt in the null hypothesis of no association. If the coefficient is negative there is a negative association and vice versa. You just need to write out the results like:

A 1 unit increase in DER is associated with a 0.008 increase in ROA. You also need to think about whether a 0.008 increase is contextually important. So if you are on diet A and your average weight loss was 0.05 kg and it was statistically significant who really cares, since A cost money and effort.

Lastly, are you still allowing each group to contribute more than one value. An assumption of linear regression (OLS) is that the observations are independent. If you have two time dependent values from the same group you are breaking a key assumption of OLS.
ok thanks a lot, i will take your sugestions in consideration.