Considering the project at hand, I will strive to follow a “Level 1 edit” which begins with revision, covers substantive editing, then copyediting, and concludes with proofreading. This method of document revision will prove most beneficial for my technical instructions as it is thorough and unanimously recommended when possessing a sufficient time schedule. However, depending on how well I progress and in what length of time, I may be forced to adopt a higher level of editing that begins at a later phase, i.e. omitting revision and starting with substantive editing.
Aside from reviewing and critiquing the documents myself, I will “take them to the people” via thorough experiments with sample readers, known as “formal usability testing” in chapter 12 of TCT. The idea is to expose the instructions to a large audience and collect as much data as possible in regard to usability. Such data would include how well the documents’ design and structure promote quickly finding information, understandability of directions, accuracy in relation to real world application and implementation, and safety. Chapter 12 of TCT presents a number of usability tests that are designed to answer the aforementioned questions. While this is true, each test emphasizes varying aspects and also differs in how it is conducted.
I am presenting my instructions in the form of a website and, therefore, retain all the advantages and disadvantages that go with that. For example, it will be considerably easier to recruit readers to sample a website and provide feedback for improvement, than distributing paper drafts. On the flip side, readers will not have the freedom for response they do with a physical document so the options for feedback are limited. To account for these, and the rest of my concerns, I plan to use a combination of usability tests, including document markup, read-and-locate test, and surveying. Together, these tests will allow readers to show precisely where they encounter difficulty/confusion, demonstrate the ease (or difficulty) of finding specific information within the document, and reviewing the overall experience and how it could be improved. I thought it was interesting that I had elected a combination of these 3 tests prior to reading “Instructor Blog #6” which emphasizes each. For this reason, I am confident that I will have little trouble composing “usable” instructions for this project, and “usable” documents in the future.
Good idea
I like the idea of using a website to gather information for your usability testing. I also see what you are saying with regards to the downsides of using a website. From your response I am guessing that you will be using a survey to solicit responses. Would it be possible to include comment boxes so that the tester could easily write in comments that didn't fit your questions? That way you could possibly regain the freedom for response that a paper survey has. I have a hard time seeing how you will implement document markup with your instructions since they will be on website. Will be you putting your website into a PDF so that it can be marked on?
Andy
All for one R9
I completely agree that combing more than one of the usability tests is better than restricting you to just one. Whatever questions you need ask or prompts you need to give in order to get the desired feedback is what you should use. Right now I have mine set as sort of a combination survey and summary but I think I might be able to benefit from using some read and locate tests as well. I think the way you plan on employing your usability tests will assuredly benefit you better than if you were to stick to just one method.
Shane
Combining Tests
Combining tests is good way to get helpful feedback to improve the usability of your instructions. My question, however, is have you considered performing different tests for each type of instruction set? Expert and Novice readers not only will have a different set of instructions in front of them (your making 2 sets for this very reason), but will also read, locate, and interpret your information differently. One test may be good at providing important feedback for your novice set, and not very much so for the expert set and vice versa.