Friday, May 18, 2012
  Search
Innovative Software for Research Reporting, online platform for survey reporting with in-depth statistical analysis for the professional expert

Survey Analysis and Reporting in the Cloud

 

Most impacting variables
Created by jdeboel on 10/26/2010 11:23:46 AM

The” Impact diagram » Correlation” analysis is used to represent the mean values of the various facets in relation to the correlation with the study’s dependent variable (sometimes multiple variables). The higher on the vertical axis, the higher the mean score.
The unique part about this graph is that it also shows the strength of the relationship (or correlation) with the dependent variable. (the horizontal axis) The more to the right, the higher the correlation.

In the graph, there is no distinction between concepts. All facets for all concepts are shown.

Below the graph you are presented with a table that lists all facets. For each of these facets you can see the concept they belong to, you can see their true value, the Pearson correlation-coefficient, and a conclusion.
 An example:
A course satisfaction study (see “DEMO” dataset) is set up by a university to find out what students think about the quality of the education the school offers.

When analysing the results using a correlation impact diagram, the researcher will conclude that in this particular survey, “Tests” scores best and “Course” and “Openness” have lesser results.(according to their vertical position) But also, very importantly, the researcher will see that “Exercises” and “Openness” have the most impact on the “Global satisfaction” dependent variable. So, one could say that since “Openness” has the biggest impact on the global satisfaction score; the “Openness”- aspect of this questionnaire is perceived as the most important aspect to help shape the global satisfaction with the school.
Seeing as the worst scoring facet has the biggest impact on the global satisfaction, one could say “the school has a big problem”. Or “the school should invest in improving the Openness facet, to very quickly improve the global satisfaction of its students”. On the other hand, the opposite could be said about the “Exercises” facet.
Correlation Graph Impact Diagram

Drobots also offers regression analysis to measure the impact of concepts. Correlation and regression analyses are a lot alike. The main difference is that regression analyses calculates the unique contribution of each independent variable to  the dependent variable; and correlation doesn’t take this into account. The “Impact diagram » Regression”  analysis is used to represent the mean values of the various concepts. The higher on the vertical axis, the higher the mean score.

The unique part about this graph is that it also shows the strength of the causal relationship (or regression)  with the dependent variable. (the horizontal axis) The more to the right, the higher the impact.

If there is more than one concept in the selected dataset, you can select the concept you want to evaluate at the top of the page. Next to “Select Dependent Concept:”.

Below the graph you are presented with a table that lists all concepts. For each of these concepts you can read their true value, the standardized OLS regression coefficient, and a conclusion.
An example:
A course satisfaction study (see “DEMO” dataset) is set up by a university to find out what students think about the quality of the education the school offers.

When analysing the results using a regression impact diagram, the researcher will conclude that in this particular survey, “Tests” scores best and “Material” and “Staff” have lesser results.(according to their vertical position) But also, very importantly, the researcher will see that “Staff” has the most impact on the “Global Satisfaction” split variable. So, one could say that since “Staff” has the biggest impact on the global satisfaction score; the “Staff”- aspect of this questionnaire (= quality of teachings, behavior of professors,…) is perceived as the most important aspect to help shape the global satisfaction with the school.

Seeing as the worst scoring facets have the biggest impact on the global satisfaction, one could say “the school has a big problem”. Or “the school should invest in improving the Staff and Material facets, to very quickly improve the global satisfaction of its students”.
 Regression analysis is used to represent the mean values of the various concepts

print
rating
  Comments
Copyright 2009 - 2012 by Kpiware - Powered by Webmove