Best of
the Lists
(including Best of BUSLIB-L)
(including Best of BUSLIB-L)
|
Best of
the Lists (including Best of BUSLIB-L) |
|
|
To be notified of new articles, sign up for email alerts or subscribe to our RSS feed |
Library surveys
The following sites offer good survey examples:
Here are the responses: Two warnings on surveys. Firstly don't expect anywhere near 100%, in fact estimate relatively low responses. Secondly it takes a while to work through the responses. We ran one a couple of years ago and while we have tried to implement some of the things that came out of it quickly, some of the others are just now being acted upon. However they are a good idea as they often tell you things that you had no idea people thought about the library and suggestions for services. It is a great idea to have a prize for participation, but one of the recognized ingredients for a valid survey is probably that it can be completed anonymously. So I would suggest that you work out a method that will allow this. It is great to use the intranet, but maybe people will have to print off the form and drop it into the library and then you could give them a raffle ticket, with no name on it. You can then draw a number at the end, but the people will have to hold on to their ticket till the draw. Also, have some paper surveys available for those who might have trouble accessing computers. I was given permission by our catering manager to leave surveys on the tables in our staff dining room, and as most people use the dining room at least once a day, I was able to reach many people. Survey questions can often give you the information you expect...and can be geared also to get the answers you want. Just look at the political survey questions and see if they aren't "begging the question." Make sure that none of the questions are two-fold, otherwise you will get an answer to only part of it and the wrong answer to the other part. Secondly, you have to somehow weight the questions when tabulating them, or some things may seem to create a "situational change" when one is not needed. In scoring such surveys one must add in some type of plus or minus for those who don't answer the survey truthfully -- just fill in the blanks sort of. Finally, if the survey points to a situation that perhaps needs to be developed, changed or corrected, move with caution. Don't jump to conclusions that the data does not support. Build in checks and balances so that in comparing the answers will result in a true picture...such as offering a chance for a written comment or explanation to the answer given. While doing my "other job" yesterday - rewriting student work in good English - I came across a useful tip in an MPhil proposal on which I was working. The student points out that the survey, as such, is a blunt instrument when seeking information. The interview allows you to explore a subject in greater depth, and to ask follow-up questions based on the answers given to your original questions. I have never yet taken part in a survey in which there was not at least one question that caused me a problem. The reason for this was usually that the survey asked for a "yes/no" answer and I wanted to say "yes but". It often happens that two people with basically the same information to give will in fact give different answers, because one will incline to one possible alternative and the other to the other! A survey cannot allow for all possible varieties of answer, or for the reasons why a particular answer is given. You can allow space for people to give reasons, but most won't take the opportunity. All this means that you have to allow considerable margins for error when working out the results of a survey. You can get more accurate information by asking people if they would be prepared to be interviewed at greater depth at some future date. The downside of this approach is that you will not get a representative sample of your respondents as interviewees - only those who have something particular to say and who are willing to say it. However, you will get a more rounded picture of things than your survey alone will give you. An alternative approach would be to carry out a few interviews before you do your survey. This will help you to see what it is that most concerns your "audience", and might help you to formulate the survey questions in ways that will produce the most helpful information I remember doing a survey a few years ago - and the question which elicited the most humorous responses was along the lines of: "What would have encouraged you to the use the library more?" More than one student answered "If I had wanted to work harder." I just saw the post about the library survey you are planning. It has been my experience that having a management sponsor send a message to your target audience, introducing the survey and recommending that the group participate, can help to improve the response rate. You might provide the sponsor with a draft message so that it is easy for them to assist you.... I always send a thank you gift to everyone who fills out the survey. I got the idea from someone else , years ago, who said they give subway tokens to customers who filled out a survey. We don't have a subway, but our cafeteria does have really good cookies - so I send a "Free Cookie" coupon to everyone. I may add a prize drawing to my survey next year - good idea! That would help increase my response rate from employees who aren't here in Denver. Before I started working here, the staff of the library conducted small focus groups to reach a diverse audience and to solicit detailed feedback about what information resources people were using and what resources they'd like to have access to. In addition to gaining some useful information, the communication served as a marketing tool for the library--who the staff was and what services were offered. Perhaps this step could be incorporated either during the formal survey or as a follow-up. A few strategies we have employed: send out on a Tuesday around 10 (on the theory that users have taken care of things that needed immediate attention for the week and will have read and processed their morning e-mail) give a week deadline send out reminders 2 days before and day of deadline, targeting just those who have not responded if possible Although you mentioned running your survey via your Intranet, I had great success using the online service Survey Monkey at http://www.surveymonkey.com Regarding your posting to the BUSLIB-L list, I thought I would send you the link to an article that I wrote for Online. See http://www.infotoday.com/online/sep02/Plosker.htm See also " Web-based surveys: changing the survey process," (First Monday, November, 2002) |
|
