Any experiment I might design to answer my questions so far will have to be a small-scale operation, working with limited resources and relatively few participants. I would likely be looking at something like a two-group RCT design with a pre-test and a post-test administered. This could take the shape of a comparison between two libraries or two library systems, one in which nothing is changed and one in which a treatment is given consisting of the implementation of an outreach program to non-readers. This might include establishing greater use of icons and picture maps to demonstrate where materials are located; and increasing materials that illiterate or low-literate adults can use, such as adult materials on a lower reading level, audio materials in different languages, magazines and adult picture books; and performing some type of community outreach.
I am honestly not crazy about this design because it presents a lot of challenges in data collection methods and establishing validity. I could probably get usage statistics fairly easily, so I could figure out whether the materials for the targeted minority group were being used, but I cannot think of any easy way to find out if the person using the materials is illiterate or has a low reading level. The only way I can come up with to identify patrons who are illiterate, short of asking every person in the library whether they can read or not, would be to somehow incorporate members of a literacy class, as Dr. Solomon suggested in his comments. I could certainly perform some sort of verbal survey on the students in order to glean their thoughts and ideas on the availability of materials for non-readers in the library. Perhaps that would give me a better idea of where I want to head with the experiment and what questions I am actually trying to answer with all this.
All this is to say, I am just about ready to learn more about flexible designs and hope that I am able to come up with something a little more sensible and practical. We'll see.
Thoughtful reflection on a possible experiment given your interest! Developing workable as well as meaningful, reliable, valid measures is (pretty much) always a challenge. Researchers often go with what is available. Literacy/illiteracy is often not an all or nothing thing (as with kids who are developing reading and associated skills, many adults have some reading skills). As you mention identification is an issue and having people self-identify by enrollment in a class (or whatever) would be a convenient way. (Are illiterate people likely to show up at a library.) While there are obstacles, they are not insurmountable. There are always tradeoffs in research. One final thought: one possible reason for using a flexible design is to better understand measurement options.
ReplyDelete