Dear researchers,
As to the my analysis in C/P effect, I need to get some attribution in my discussion section. Due to the extent conceptional framework for the culture it is hard to settle a causal relationship for it containing firm performance. Just for your information, I used Denison's Organizational Model, employed cross sectional OLS model and my dependent variable is ROE. There is a slightly relation between mission trait and ROE. Except this there is no significant relation in other traits.
I beg your your empirical recommendations in this issue.
Thanks in advance,
Adem.
Hi Adem
Are you able to provide your model with the dependent variables? Perhaps your proxies that relate to the model are not suitable.
Dear professor Cox,
Thanks for your participation.
I can not properly understand what you mean but I think it's all about a goodness of fit issue.
I firstly would like to mention that my data is linear and I'm using OLS estimator for this. At first, I intented so many dependent variable seperating, classifying and organizing them profitability, growth, working capital, capital structures and market based variables (total 14 variables, TQ included). But it was too hard to estimate each of them is selected as dependent variable.
So I decided to select just one dependent variable and set up my design over it.
Later on, I added ROA and debt to equity ratio, then SIZE and AGE to my regression model. This helped me to acquire a base regression model.
These were all significant and goodness of fit for the model were increased in all way. Finally added independent organizational culture variables; Involvement, Consistency, Adaptation, Mission and OC itself. That's no trouble here.
My data is cross-sectional because of the OC data. I picked them up via an OC survey and these valid for just one year as the financial variables are estimated average of 5 years and not contain of extraordinary incidents.
So, I tiled up my research method as well but I think there is a missing point over OC assessment and could not connect it to the results. This is rooted in OC itself, I think.
Dear Adem
I suspect the reason that the organizational culture variables are (in general) not statistically significant in your regression model due to the other independent (financial) variables. That is, if your dependent is ROE and among your independent variables you have profitability (like ROA) and capital structure (like Assets/Equity) then basically the right-hand side of the equation is a definition of the left-hand side. As ROE = ROA X A/E this is generating an extremely high R-squared leaving little explanatory power for other variables in particular your organization culture variables. Another approach to determine the efficacy of the organization culture variables explaining ROE is to conduct a univariate t-test prior to running your OLS regression.
Dear professor Cox,
Your analytic sounds logical and worth to try.
Thank you very much.
Adem.
- Badges
- Science topic
- Similar topics
- Methodology
- Research Methodology