Information Outlook, Vol. 6, No. 7, July 2002
If You Don't Know, Ask: The Art and Craft of Survey Development and Analysis
by Gloria Dinerman
Gloria Dinerman is president and founder of The Library Co-op, Inc., a national library service and consulting corporation specializing in projects and automation assignments where customization is essential to success.
Building Your Survey
In the good old days before the extensive use of automation, when a survey was mailed out, an expected rate of return was about five percent of the targeted group. More extensive participation was a pleasant surprise. Even with surveys that include self-addressed stamped envelopes and phone calls as follow-up reminders, a return rate of 25 percent was an accomplishment.
Today, with desktop computers now standard in every office, online surveys produce a far higher percent of responses. This leads to a more accurate analysis of the data received.
Although survey construction is customized to the environment of the institution or corporation needing the information, there are a few general commonalities that apply to all surveys, regardless of the entity requesting the data.
· Precision of the wording is vital to getting the correct information. "Have you ever used the library?" is much too vague. "Do you use the library regularly?" is also too vague. Try this:
Do you generally use the library:
once a month once a week every day less than once a month
· "Do you usually find the information that you need?" is not as precise as:
Do you generally find the information that you need:
in a print source in a non-print source online
· The most accurate information comes from:
Internet sources q aggregated databases q professional journals q
State the Purpose
In order to get precise information, the purpose of the survey should be stated to the respondents at the top of the survey form.
The following examples reflect some of the reasons a library may want to conduct a survey:
· to find out what type of material is being used the most or least;
· to find out what your users need/want;
· to find out if your service has been satisfactory;
· to find out where you can expand your service;
· to justify the need to retain an information center;
· to justify your budget;
· to justify the need for additional personnel;
· to justify the need to weed out old material that is no longer relevant to the company or institution;
· to justify the need for expansion;
· to question whether or not your information center should become virtual; and
· to question whether or not your budget should be reallocated.
The survey can be beneficial to librarians in both academic and corporate settings.
The Corporate Survey
With more and more end-users doing their own research; with online taxonomies getting more expansive and user-friendly; with less corporate space being reserved for hard copy; with corporate accountants constantly looking for ways to save moneyit becomes clear that corporate libraries have to justify their existence with hard proof that they are essential to the corporate structure. One of the more successful ways of proving the need for an information center is by having the library users respond to a survey that not only stresses their reliance on the library services, but also the importance of having professionals on staff. A survey of this type should be done before any administration formulates plans to eliminate the information center. In most cases of downsizing, the librarian is the last to know that the job is being downsized and because of this gap in communication, it is prudent to have information on hand that confirms the importance of the library.
Corporate personnel are extremely busy and do not have the patience to respond to any questionnaire that takes more than five minutes to complete. If the survey is too lengthy, the respondent may give up in the middle and invalidate the whole questionnaire.
Here are some sample questions:
· Are your reference inquiries responded to in a timely manner?
Always Generally Sometimes Never
· Do you request an analysis of the retrieved information by the librarian?
Yes No
· Do you request an abstract of the retrieved information?
Yes No
· If an abstract is requested, is the written material presented concisely and correctly?
Always Generally Sometimes Never
· Does giving your research problems to the librarians save you time?
Yes No
· Do you feel the information center is cost effective for the corporation?
Yes No
· If your company did not have a research center, you would get your information from:
Using Internet yourself
Using aggregated databases yourself
Using reference material in the public library
Other professional colleagues
Would not get research material
· Is the librarian alert to new search techniques and new aggregated databases?
Yes No
This type of questionnaire gives you a foundation to prove the value of your work. Your supervisor should be presented with an analysis of the results along with the departments that responded and the number of people within the department who responded. These statistics are easily analyzed using an Excel spreadsheet, but any spreadsheet program should be able to produce the quantitative results.
Academic Surveys
An academic institution can have many libraries and many departments, special collections and archives. The most needed information is coordinated with the professors who want the students to have reference material relating to their classes. If the availability of these references is not adequate to satisfy the research requirements of the courses, then everyone involved becomes dissatisfied.
To find out if the library is adequately serving the needs of the university, some institutions regularly put a questionnaire at the checkout or reference desk, asking:
Did you find your material?
Did the librarian assist you?
Did you use print sources or electronic sources?
Are the library hours adequate for your schedule?
Are you always, or almost always able to get a computer when you need it?
Have you ever gotten material through interlibrary loan? Did it come in a timely manner?
Do you use our special collection material? What subject?
If your library uses self-checkout after hours, does it cut down on the loss of material?
| A Summary of the Survey Process: | |
| · · · | Select your inquiry or theme |
| · · · | Construct your questions |
| · · · | Select your analyzing database |
| · · · | Pretest the survey |
| · · · | Conduct the survey |
| · · · | Enter the completed data |
| · · · | Tally the responses |
· · · | Interpret the results |
| · · · | Write your conclusions |
Demographic information is more important in an academic survey than in a corporate one. The main components of demographics are age, ethnicity, education and economic level. Age and economic level are always given in ranges, i.e.:
under 18 19-40 41-60 Over 60
Under $30K per year $30K-$50K per year
$51K-$75K per year Over $75K per year
The outstanding difference between the results of a corporate library survey and an academic library survey is that in a corporate survey, the results may have a direct effect on your job retention. The bottom line of a corporation is pulsating with income figures and the library is a cost factor. In a university, the measurement of service is taken by depth of resources and response time. Academic librarians get a lot of attention from vendors because of their buying power.
From Library Journal (9/1/01), "Moving from Books to Bytes," by Andrew Richard Albanese:
"LJ's 2001 surveywhich gathered data from a balanced sample of libraries at research institutions, universities, four-year colleges and junior colleges nationwideonce again reflects the considerable spending power of academic libraries. In 2001, the median materials budget for a university library was roughly $565,000 compared with $174,000 spent by four-year colleges and $68,000 for junior colleges. Universities added 6,831 new monographs on average; four-year colleges added 3,095 new titles and two-year schools added 1,643. In contrast, the more than 120 institutions comprising the Association of Research libraries (ARL) operated with budgets closer to $6 million than $600,000."
None of the above would be known without the use of a survey. And as seen in these statistics, the purpose of the survey was to show the differences in the library expenditures in various types of institutions. You do not find that type of comparison in corporate libraries.
In the academic library, the preparation of the survey is geared to whether the institution is interested in statistics in relation to internal procedures or external comparisons. Rarely is an academic survey constructed without the assistance of a committee. Academicians like to think as a group and each decision is arrived at by consensus. Sometimes the committee gets bogged down in minutia and it subverts the progress of the finished product.
Relevance of Statistics
· What is the number of regular patrons (employees who use the library once a week or more)? If one person from a 12-member department responds, it is less meaningful than if five members from a nine-member department respond.
· You want to know the percentage of favorable or very favorable comments, by department and by the total number of respondents.
· You want to know if there are any weaknesses in your service.
· Is your collection relevant to the product or services of your company or has part of the collection become obsolete?
· Are you responsive to the needs of the researchers and are they providing you with changes in focus or product line that will allow you to regularly improve your services?
· Do you offer training and orientation to new employees?
Since the Internet is expanding in its number of available sources and since it has become more user-friendly, have your reference requests been:
Increasing Decreasing No difference q
If yes, by how much?
10% 20% More than 20 %
If you plan to use the survey in a formal presentation, it is suggested that you format it using a split screen. The left side of the screen will show the questions and the right side will show the totals of the answers. The final set of screens can be converted into graphical results.
The single most serious challenge for the corporate librarian is convincing the budget managers not only to acknowledge the results of the survey, but also to act upon the conclusions drawn from the responses.
Without the benefit of administrative support it is not likely that any credence will be given to the results. This administrative support has to be developed early in your career. Every compliment, every credit, every award, every letter of commendation should be brought to the attention of the supervising staff person that has anything to do with the future of the library. A survey can prove your worth, but only you can open the channels for action.
Wrap-Up
While surveys can be very helpful, it is important you remember some key points before diving in.
It is recommended that you conduct a pre-testing or sampling of your survey using a total of 10-15 percent of the respondent pool. This is to ascertain that both the delivery methods and the database are working correctly. The unexpected problems that surface can be corrected before the survey is actually conducted.
Since the survey should be kept short (10 questions maximum), the person creating the survey should write in simple sentences and ask for only one item of information per question. Evaluation is even more important than the writing itself, and responses must be quantifiable.
If you follow these rules, your survey can be very valuable because surveys give objective responses to subjective questions. There is no right or wrong responsejust an assessment of a condition that already exists. However, these opinions have a basis of fact and stand to serve as a foundation for improved staffing, increased collection and evaluation of existing facilities. It is very difficult to judge ourselves in an objective light without some statistical support.
RSS
Feedback form
