Heuristic Evaluation of Prototypes

From CS 160 Fall 2008

Jump to: navigation, search

Contents

Heuristic Evaluation of Prototypes (Individual)

Due: Friday, Oct 31, 2008

Goals

The goal of this assignment is to learn how to perform a heuristic evaluation of a real user interface.

Overview

You have been hired as a consultant to another group in the class. They are building a new user interface for their course project, but they would like some outside assistance in finding some problems with their prototype interface. The group you evaluate should be (your own group + 1) (mod 9). Since most groups arent actually using numbers, the sequence is

Group:Lucky Seven
evaluates
Group:Phi-tus
evaluates
Group: Group Ate
evaluates
Group:一三三七 (One Three Three Seven)
evaluates
Group:!Xobile
evaluates
Group:TBD
evaluates
Group:Orquesta
evaluates
Group:4
evaluates
Group:BuTtErFlY
evaluates
Group:Lucky Seven

Evaluation

You will perform a heuristic evaluation (on your own) of their user interface using only the materials they turned in for the last project report. Using their tasks, scenarios, UI design, screen shots, and interactive prototype you will apply Nielsen's heuristics to the UIs. You should be able to get all of this information from their web page. Read their report first and then examine their prototype (i.e., run it). Your evaluation will use both the information in the written report and the prototype.

Please use the set of heuristics I gave in class (also described in Nielsen's chapter as the H2 series) and the numbering scheme from class (e.g., H2-1, H2-2, etc.). After finding all of the problems, go back and assign a severity (from the 0-4 scale defined in class) for each violation you found. You will produce a report showing the problems in the interface.

Reminder

Obviously, for other students to evaluate your prototype, you should make sure you submit it ON TIME. Your group will definitely lose points for late submission on the interactive prototype assignment. It’s also important to leave clear instructions for other students to be able to load and run your application (and make sure they are accurate).

Report

Your report will list each of the problems found in the following format:


problem# [heuristic violated][severity]

description of problem and reasoning why you think this violates the heuristic


For example:

5. [H2-4 Consistency & Standards] [severity = 2]

The interface used the string "Save" on the first screen for saving the user's file, but used the string "Write file" on the second screen. Users may be confused by this different terminology for the same function.


6. [H2-3 User Control & Freedom] [severity = 3]

The interface brings the user into a set of preference screens when they select "New User", but doesn't allow the user out of the dialog until they fill out all 4 screens. There is no way to cancel from any of the screens if a user came into the first screen by accident.


Your report will also summarize the number of violations found in each of the ten heuristic categories and give the total number of violations in the entire interface. You should also list the total number of violations in each severity category.

Finally, your report should close with some overall recommendations you have for improving the user interface given what you read of their description.

Deliverable

Your printed write-up (turned in on paper) should follow this outline with separate sections for the top-level items:

1.Problem (one sentence description of the UI you are evaluating)

2.Violations found

3.Summary of violations

4.Recommendations

Grading

You will be graded on how complete is your HE in terms of coverage of the presented user interface design, clarity of your violation descriptions, and quality of your recommendations. You should concentrate on the interface the group has designed, not only on what has been implemented. Reports that continually report on features that are missing, but will clearly be added will be marked down (e.g., "there should be help on this screen... and this screen..." -- if it is a globally missing feature you can report it once). We want to focus on evaluating what they have designed so far.

Personal tools