Organisational Development Assignment

Guide to Statistics used in the Assignment

Means and Standard Deviation

For each variable included in the case study (e.g., LMX, organisational justice) a mean (i.e., average) and standard deviation is provided. The mean tells you the average rating provided by employees in each department for each variable.

So, for instance, for the mean rating for meaning in the Intelligence Products department is 1.50. This tell you that, on average, employees perceive that their job meaning is 1.5 out of 5. Being an average, some employee will rate higher and some lower, but overall you can conclude that in general, meaning is low in this department. As a rule of thumb averages between 1-2 out of 5 would be considered low, between 4-5 would be high. Between 2-4 is somewhere in the middle.

As well as the mean, each variable also has a standard deviation (SD). The SD is a measure of how much the members of a group differ from the mean value for the group. A large SD suggests there is a great deal of variation in peoples’ responses to a scale. For example, the mean LMX score for the Software Development teams is 2.23, with a SD of 1.66. This is a large standard deviation and suggests that there is a lot of variability in employees’ ratings of LMX. On a 1-5 scale, a SD of between .10-.50 would be considered small (suggesting general agreement across people in the group), a SD of .50-1.00 would be moderate (suggesting a moderate amount of variation in score) and over 1.00 would indicate a relatively large amount of variation in how people rate that variable.

The mean scores on variables can be used in your assignment to help provide some evidence to support your arguments. You do not have to use them, but it might be helpful. For example you might argue that a lack of meaning is a big issue effecting the intelligence products department. This is evidenced by the low mean score on that variable.

Also included in the assignment is a correlation table. Again you do not need to use this if you don’t want to. Correlations give you some information about the relationships between the variables measured in this organisation. Specifically, correlation is a statistical technique that can show whether and how strongly pairs of variables are related. For example, height and weight are related; taller people tend to be heavier than shorter people. 

Perfect positive correlation would be 1.00, perfect negative correlation would be -1.00. A correlation between .10 and .20 generally represents a small relationship, so as one variable increases so does the other, but to a small extent (i.e., there is a lot of variation). A correlation between .20 - .40 is a moderate relationship. Over .40 would be considered large. So, for example, in the assignment there is a correlation of -.50 between LMX and turnover intention. This negative relationship suggests that as scores on LMX increase, turnover intentions decrease. Thus, LMX is negatively associated with turnover intensions. Put differently, the higher the quality of the relationship between leaders and follower, the less likely employees are to want to leave the organisation.