Programming Assignment 2

CSC423/324 - Data Analysis
Due: 11/10/2003

The pervasiveness of mobile communication technologies has led to considerable recent interest in methods of efficient text entry for these devices. Your colleague is interested in this issue and has asked you to help with a usability experiment that compares two different methods of text entry for mobile phones. The two methods are the 12-Key Keypad text entry method and the T9 text entry method.
Note: For more details on usability see Jakob Nielsens useit.com site. In particular, the Usability Metrics essay, the Novice vs. Expert Users essay, and the Graceful Degradation of Scalable Internet Services essays are relevant and interesting.

12-Key Keypad Method:

Each key must be pressed one or more times by the user to specify a character. For example, pressing the 2 key, results in the character a. Pressing the 2 key two times, in rapid succession, results in the character b. Pressing the 2 key three times, in rapid succession, results in the character c.

T9 Method:

Uses a built-in dictionary to look up words for a key entry sequence. For example, to enter the character sequence abc, as above, the user would press the key sequence 2-2-2. Similarly, for the word phone, the user would press the key sequence 7-4-6-6-3. The T9 method uses the key sequence entered, and its built-in dictionary, to identify word possibilities. Where multiple words are possible for the entered key sequence, T9 retrieves the most common word. If necessary, users have the option of scrolling through the list of candidate words.
Note: For comments on mobile communication issues see the paper Designing the next generation of mobile communication by Christina James the chief designer of the T9 method.

Your colleague explains that she believes that the T9 method is the more efficient method. She further explains that to investigate this issue, she selected two random samples of novice mobile phone users. One group was trained to use the 12-Key Keypad method and the other group was trained with the T9 method. Each user was allowed extensive practice time over several days. The users were then observed completing a standard task suite that involved 3 tasks. The time required to correctly enter the text specified for each task was observed. Your colleague points out that by measuring time until correct completion, both speed and accuracy are taken into account.

You have been provided with this data file. Each observation consists of the following values:

• Task #1 (seconds); 1-3
• Task #2 (seconds); 4-6
• Task #3 (seconds); 7-9
• User ID; 10-16
• Observer ID; 17-21
• Satisfaction Score; 22-23
• User Group; 24 (i.e. T - T9, K - 12-Key)
• Status; 25 (i.e. I - Incomplete, C - Complete)

Notice that the times provided are for the individual tasks in the suite. You are interested in total task time. Notice that a Status indicator has been provided. You are only interested in observations where Status is C. Several users did not complete the task suite and so the task times for these users should be ignored.

Note: Do not edit the data in any way. You must code your SAS program to read each observation as defined above and ignore those observations that are not needed.

Conduct a thorough analysis of these data. You will need to conduct a test of hypotheses and submit a report summarizing your findings. See additional details below.

1. Write a SAS program to analyze this dataset. (50%)

   Your program should do the following:

1. Read your data from an external file.
2. Compute the total task time for each observation.
3. Analyze observations where Status is C only.
4. Execute the PRINT procedure.
5. Use the appropriate SAS procedures to produce the p-values needed to conduct your hypothesis test.

Note: For PROC PRINT, be sure to use labels for column headings rather than variable names. Use names for data sets and variables that are meaningful. You should generate an appropriate title for the output of these procedures.

2. Write a short analysis (no more than two pages) of the output of your SAS program. (50%)

   Remember that your analysis is a test of hypotheses and so should at least address the following:

1. State the primary hypotheses. That is, the NULL and ALTERNATIVE hypotheses for the experiment described above.
2. Address all conditions that must be established prior to assessing the p-value for your primary hypotheses.
   Note: That is, do you need to establish normality, or do you need to establish equality of variances, in order to address your primary hypotheses? You may make the equal variance/standard deviation assumption but do not assume normality (i.e. state and address the normality hypotheses).
   Extra Credit: Show how you would use SAS to address the equal variance/standard deviation issue. That is state the equal variance/standard deviation hypotheses and address them using the appropriate SAS procedure (see the example problem in these lecture notes).
3. Determine the p-value for your primary hypotheses and comment on the significance of this p-value.
4. Given your findings, and considering the issue being investigated, briefly discuss your findings.