Magazine article American Libraries

Great American Public Libraries: HAPLR Ratings, 2000

Magazine article American Libraries

Great American Public Libraries: HAPLR Ratings, 2000

Article excerpt


Not all libraries will greet this new edition of Hennen's American Public Library Rating (HAPLR) index with joy. Some will say that the rating system is flawed, or that public libraries vary too much to assign meaningful rankings.

But others will be happy to see the comparisons again. Those in libraries that rank well are proud of their achievements, of course; but many in moderately or poorly ranked libraries are happy to point to their low ratings as further proof of the need for additional funding.

This edition of the index is based on 1998 preliminary final data from the Federal-State Cooperative System (FSCS) as published on the Web in June 2000. Has much changed in this edition? For better or worse, the answer is not much.

* Usable national data is still not available on Internet use, electronic access, or building size; however, progress is being made on data collection in these areas. The ratings have been criticized for not considering these factors, but the raw data is not collected nationally using consistent methods.

* The first HAPLR index (AL, Jan. 1999, p. 72-76) divided libraries into just four population categories. The current index and the previous one (AL, Sept. 1999, p. 64-68) divided libraries into the 10 population categories used by FSCS.

* Of the approximately 9,000 public libraries in the United States, about 1,000 do not report annual visits and another 1,000 fail to report reference queries answered. Consequently, FSCS imputes the data. "Imputing" means to guess, using statistical principles. The current HAPLR index cannot use this data because it is not yet available; thus 2,000 libraries are left out of the ratings. (The second index relied on imputed data and thus included 9,000 libraries, whereas the first strictly used unimputed data and covered only 7,000 libraries.)

In the 21st century, libraries that still do not track visits and reference activities are strongly urged to do so. Many libraries that ranked well when these factors were imputed have fallen off the radar screen because they cannot or will not report these elementary data elements. When FSCS releases its final report with the imputed data, the HAPLR database will be adjusted accordingly.

The structure of the HAPLR index

The HAPLR index uses six input and nine output measures (see the table on page 51), calculated from FSCS data. Each factor was weighted and scored. Then the scores were totaled for each library within a population category to develop a weighted score.

Responding to critics

The previous index met with many objections and criticisms, and I have tried to address some of them here.

Insufficient number of criteria. Adding more data categories to the index is possible. Program attendance, video circulation, and children's circulation are a few of the measures that have been suggested. However, I have resisted adding more in order to retain consistency from one version to the next. Weightings and categories remain the same as in the previous index.

Flawed methodology. Jim Scheppke in "The Trouble with Hennen" (Library Journal, Nov. 15, 1999, p. 36-37) misstates the HAPLR methodology. The scoring and ranking is within each population category, not across all categories as Scheppke indicates. The comparisons and rankings in each category, from expenditure per capita to visits per hour, are only to libraries of comparable size, not to all 7,000 libraries. Scheppke also notes that professional judgments would severely downgrade several of the top-rated libraries in Oregon because they lack adequate building space. He fails to add that neither Oregon nor FSCS collects the data on building space necessary to add this dimension to a rating system. (That situation will be remedied in the near future. Partly because of the HAPLR ratings, FSCS will soon begin to collect building data. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.