# Evaluation Services

UTS offers evaluation services to any Department or College on campus.

To use our service you will need to purchase forms for the evaluations to be recorded on from UTS. These are the same forms used for test scoring and can be purchased by emailing scores@uga.edu. The form (229633) is used for evaluations and research surveys.

Images of Forms:

The only information gathered from this form for evaluations are the Identification field and the responses in fields 1-100

Below is a sample of the printed reports with information on how to read the reports that UTS provides for Evaluations.

This sample set of evaluation forms had 43 respondents total. There were five questions that each had five choices.

There were five separate sets in the evaluation sample, subset data is separated by filling in a left justified common number in the ID field of the answer sheet. Each Instructor, Lecturer, TA or class being evaluated is assigned a different number so that this data has a separate printed report that matches the number assigned. For best results, these numbers should be filled in before students are given the answer sheets to record their responses. This method clears up most errors in the reports. These numbers need to be a minimum of two digits long.

In Evaluation Report 1, you have the data for the entire set of forms for evaluation. This set is commonly referred to as “totals” since the field at the top of the page after “EVAL 2009, Code:” is blank. The report gives the average of all forms scanned and is often used to compare the percentages of the various sections in the set.

“Number of respondents” is 43; this is the number of forms scanned for this set. This evaluation set had five questions for the respondents to answer each with five choices.

At the top left hand of the page you will see “Question Number”, just below this are the results for the five questions labeled 1-5.

Across the top

(----- 1-----) (-----2-----) (-----3-----) (-----4-----) (-----5-----) (- OMITS -)

N per N per N per N per N per N per

These numbers (1-5) show the data for the choices of the questions in the columns below.

“N” stands for number of respondents who choose this answer “per” is the percent of respondents that chose this answer.

On the top right you will see the Mean calculation below the word “Mean” and next to that the “Standard Deviation”.

These are calculated using the formula below.

Please see the formula below used in our program for standard deviation. It is divided up into 3 steps. I also plugged in the numbers for Question #1 from the Response Summary dated 05/11/05 to calculate the standard deviation. We can discuss further if needed.

Al Hardy (Many Thanks to Al!) translated the formula from the program for us.

1. Compute Mean. In the case of Question #1, the Mean is 1.34.

2. Compute Variance:

Variance = (1 - Mean) x (1 - Mean) x (#1 occurrences) +

(2 - Mean) x (2 - Mean) x (#2 occurrences) +

(3 - Mean) x (3 - Mean) x (#3 occurrences) +

(4 - Mean) x (4 - Mean) x (#4 occurrences) +

(5 - Mean) x (5 - Mean) x (#5 occurrences) divided by the

Total occurrences for question #1 (126).

= (1 - 1.34) x (1 - 1.34) x (92) +

(2 - 1.34) x (2 - 1.34) x (28) +

(3 - 1.34) x (3 - 1.34) x (4) +

(4 - 1.34) x (4 - 1.34) x (1) +

(5 - 1.34) x (5 - 1.34) x (1) divided by 126.

= .1156 x 92 +

.4356 x 28 +

2.7556 x 4 +

7.0756 x 1 +

13.3956 x 1 divided by 126

= 10.6352 + 12.1968 + 11.0224 + 7.0756 + 13.3956 divided by 126.

= 54.3256 divided by 126 = .43115555

Variance = .43115555

3. Compute Standard Deviation = Square Root of Variance

= square root of .43115555

= .65662435 or .66 rounded

So the Standard Deviation for Question #1 is .66

In Evaluation Report 2, the Space after EVAL 2009, Code: is blank. These are called errors. This is due to the ID numbers not being bubbled in. The program cannot attach this set of data to any of the subsets.

In Evaluation Report 3, the results for the subsets with the correctly filled in ID numbers that identify the separate sections to be evaluated.

If you add the totals after “Number of respondents” along with the error you will have the total number of respondents as seen in the first report (Evaluation Report 1).

Please call or email Carol Alexander if you have any questions about the evaluation services that we offer. Training sessions are available for any departments interested in using our service.