Denison Model
Glossary
Beliefs and Assumptions
The underlying, unwritten, taken-for granted beliefs, perceptions, thoughts and feelings that are the ultimate source of values that guide employee behavior.
Percentile
A percentile is simply a percentage. This method classifies your organization's scores as a percentage score compared to the average scores of other organizations (this average is called a norm). A familiar example of a percentile would be when your child visits the pediatrician. If your child weighs 10 pounds, 6 ounces at the 2-week check-up, the doctor could tell you that they are in the 92nd percentile. This means your child weighs more than 92 percent of all other children of the same age.
Quartile
A quartile is simply a broader, more general way to look at your percentiles. The 1st quartile represents the 1st to 25th percentiles, the 2nd quartile represents the 26th to 50th percentile, the 3rd quartile represents the 51st to 75th percentile and the 4th quartile represents the 76th to 99th percentile. Following the same example as above, if your child's weight falls into the 92nd percentile, then that would also be the 4th quartile.
Norm
A quartile is simply a broader, more general way to look at your percentiles. This is what your results are being compared against. Currently, over 700 organizations are included in the norm for the Organizational Culture Survey. This means that your results are being compared against the other 700 companies. The database is updated frequently, and we are constantly expanding our norm base. The "norms" for each item and index are established by re-computing the "cut-points" for the 25th, 50th and 75th percentile each time the database is updated. The database includes a wide variety of large and small businesses from around the world, including manufacturing, service, retail, financial, high-tech, non-profit and government organizations.
Index Scores
A quartile is simply a broader, more general way to look at your percentiles. The An index score (Strategic Direction & Intent, Goals & Objectives, etc.) is computed as the average of the line items' raw scores, which is then compared to the norm for all organizations on that index. It is NOT just an average of the line items. The raw average is computed by taking the scores from the surveys (1-5) and calculating each item's average. For example, let's say that the raw average of the 5 line items that make up an index are 3.4, 3.6, 4.0, 4.1 and 4.2. To compute the index score, we 1st take the raw average of these 5 items (which is 3.86), then compare this average (3.86) against the norm for all organizations for the same index (their raw averages). Since this score is determined from the raw averages of the line items instead of an average of their percentiles, you cannot just take an average of the line items' percentiles for the index score.
Line Items
A quartile is simply a broader, more general way to look at your percentiles. The The line items are the individual questions from the survey, grouped by index. Percentiles are calculated for each line item by computing the raw average of the item (adding up the responses for that item and dividing it by the number of people answered the question) and then norming it against other organizations to create the percentile score. They are also color-coded to match the Denison Model (the Involvement pages are printed in green, the Consistency pages in yellow, Adaptability pages blue and Mission pages red). You will find a "+" and "-" sign next to the 5 highest and lowest scores. These scores will also be listed on the Highest and Lowest Scores page in the Summary Report (see the description below for more information about the Highest and Lowest Scores).
High/Low Scores
This page shows the 5 highest scores (highest score 1st, 5th highest score last) and 5 lowest scores (lowest score 1st, 5th lowest last). If there are 2 or more scores that are the same, the scores go through statistical analysis (called a z-score) to break the tie. For example, if your highest scores are 92, 90, 89, 88, 87 and 85, but you have 2 line items that are in the 85th percentile, a z-score analysis is run to determine which of the 85th percentile scores is statistically higher, and that score is then used.
N
This is the number of people that took the survey.