Subscribe: Comments for Lib(rary) Performance
http://libperformance.com/comments/feed/
Preview: Comments for Lib(rary) Performance

Comments for Lib(rary) Performance



About library statistics & measurement - by Ray Lyons



Last Build Date: Mon, 19 Dec 2016 23:02:54 +0000

 



Comment on Statistical Regression by Ray Lyons

Mon, 19 Dec 2016 23:02:54 +0000

Interesting. Thanks so much for that information, Joyce. I believe I used Compare Your Library in 2015 and it was intact. But my recall can easily be off by a year at least! Also in the past year or so I tried to access PLS data interactively--either the Compare Your Library tool or perhaps the IMLS Data Catalog--and got an out-of-service-due-to-security-issues warning. I presumed due to suspicions of hacking of federal sites. Sorry for my poor recollection.



Comment on Statistical Regression by Joyce

Mon, 19 Dec 2016 18:26:10 +0000

As I understand it, new tools had to be created because the old tools were actually developed by and belonged to the Census Bureau. The states used to work with the Census Bureau staff to review and prepare the data before it was finalized and sent to the IMLS. When the IMLS and Census ended that contract/relationship, the public-facing tools all disappeared along with the Census Bureau.



Comment on That’s the Wrong Question by Concerned Citizen

Wed, 13 Jul 2016 14:35:50 +0000

I am sorry that I didn'f find this post sooner. Thank you for the insight you bring to the situation involving the Cuyahoga County Public Library. You understand the heart of the matter. CCPL asked citizens on one occasion IF they wanted a new library. As you note in your blog,citizens said no--actually I think it was 98% or 94% of respondent said they did not want a new library. CCPL continues to this day to use bogus "surveys" to justify new buildings. You are correct. The library asks people what they want in a new library, or what services do you wish your library had. They do not ask DO YOU WANT A NEW LIBRARY? CCPL also said that the new buildings would save money--but they have never proved this or tried to do this. CCPL said they had to sell a building that is on the National Register of Historic Places and is an Ohio Landmark because it needed $5 million in repairs. So, they replaced it with a $13.6 million new building. Keep up your insightful posts--I am subscribed now.



Comment on Do No Quantitative Harm by Linda Anderson

Mon, 26 Jan 2015 22:49:12 +0000

Yes, you hit the nail on the head several times here, obviously this one - "It must be quite a surprise to university teams who pored and sweated over proposal drafts...to learn that libraries feel justified in claiming from 80% or 110% credit for the teams’ work."



Comment on Roughly Wrong by raylyons

Sun, 12 Jan 2014 22:35:15 +0000

Incidentally, the IMLS website has a page entitled Research Guidance with this link to the Office of Management & Budget Standards and Guidelines for Statistical Surveys. For federally funded research, certain nonprobability sampling methods are permissible if researchers document the need and "are able to measure estimation error." However, convenience sampling is not a permissible method under these standards and guidelines (because it leaves researchers in the dark about estimation error).



Comment on Roughly Wrong by raylyons

Sat, 11 Jan 2014 16:24:39 +0000

Thanks for your comment, Seth. It is the Urban Institute that recommends surveying the entire population of clients in public agencies. They note that this will only be feasible for smaller agencies. I have no examples of public libraries using a "sound sampling strategy," by which I mean drawing representative samples, more officially known as "probability samples." My impression is that libraries follow the lead of larger organizations like ALA, AASL, OCLC, and the University of Washington iSchool Impact Study, all of which post links to their surveys on library websites. Thus, all of these surveys obtain only self-selected (convenience) samples. You can get a general idea of how sampling fits into survey research methods from the American Statistical Association's primer on surveys and in Basic Research Methods for Librarians by Connaway and Powell. Actual steps for drawing a simple random sample appear at Dissertation Statistics. Methods for drawing other types of probability samples will appear in most survey research text books.



Comment on Roughly Wrong by Seth

Fri, 10 Jan 2014 20:12:56 +0000

You suggest that surveying an entire population ( all of the users of a library) would be better than a sloppy convenience sample. I agree, but some times a survey of the population is not feasible so you have to resort to sampling. Can you give some examples where a library survey was done that used a sound sampling strategy? It would be helpful in planning a library survey to have some practical guidance in selecting a sample.



Comment on A Hard Row to Hoe by From Ray Lyon's blog ... about LAC and Q&A on validity of inferences | Library Assessment

Mon, 02 Dec 2013 18:17:55 +0000

[…] clipped from libperform.wordpress.com […]



Comment on Statistical Hearsay by Opdracht 3 | Sarah El Haouari

Thu, 10 Oct 2013 09:52:49 +0000

[…] https://libperform.wordpress.com/2012/10/13/statistical-hearsay/ […]



Comment on Quadruple Your Statistical Knowledge in One Easy Lesson by Catherine

Tue, 20 Aug 2013 01:27:57 +0000

Thanks for this encouraging entry ! Unfortunately so many libraries rely on self selected surveys and pay money to have them collected and analysed.. without ever understanding the problems with this.