Service Quality
Improvement Council
Assessment Group
Minutes, September 21, 1999
Martha has completed the draft of the statement of ARL definitions. Please
send her your comments.
Will reported on sampling. Key to any sampling plan we develop is that the
data collection be randomized. A random set of dates can easily be generated
using the RAND function in Excel. Along with randomness, we may want to consider
in the future a sampling method that will stratify by library unit, or by
time period.
It will also be important to try to gather information on all reference
transactions. This means gathering data on transactions happening away from
the Reference desk. Email reference questions must be included in the sample.
We have not yet determined whether email will be counted on the day it is
sent or the day it is answered.
Our proposal will be to sample on two days each month. The population sampled
from will be all days the library is open and providing service. This means
possible days to be sampled from must exclude seven days: Christmas, Christmas
Eve, New Year's, New Year's Eve, Thanksgiving, Memorial Day, and July 4th.
A possible problem with this approach is that some of the libraries (Medical,
for example) do not deliver reference service on weekends. If the two sample
days happen to fall on weekend days during one month, these libraries would
report zero reference transactions. This should be balanced out over the
year. The results of our sampling will have to be evaluated at the end of
the year.
There will be some checks at the end of the year that will help us with the
evaluation of the sampling system. Sterling will continue to keep a full
count of all reference transactions. The Medical Library will also continue
to count reference transactions on its own sampling days in those months
when the ARL sample falls on weekend days. This will provide data for comparison
with the samples.
The department heads currently responsible for collecting ARL statistics
will continue to collect the statistics under the sampling system. An email
message will be sent to yulib-l on sampling days, reminding everyone to keep
track of reference transactions. This data will be handed over to the appropriate
department head. The department heads will also receive an email message
reminding them to gather the statistics from staff and to submit the data
to AUL for Access Services by the tenth of the following month. The email
message will contain a form for submitting the data. The data they submit
will simply consist of four numbers: reference transactions in person, by
phone, email, or other.
Rich will write the two email messages and the form. The committee did not
decide whether we want to create a form for actual data collection.
Recorded by Katie Bauer,
kathleen.bauer@yale.edu
