Главная >> Ресурсы >> Журнал "Научные и технические библиотеки"
Все статьи, опубликованные в рубрике: ПИСЬМО В РЕДАКЦИЮ журнала "Научные и технические библиотеки", начиная с 2013 года
Год: 2014, Номер: 08 Sukiasyan Eduard.
Сведения о данной статье в Google Scholar
Год: 2018, Номер: 02 Sukiasyan Eduard.
Сведения о данной статье в Google Scholar- Show annotation ▼
We are introducing the “Model standard of the public library”, where I once read that "scores can be expressed from zero to ten". Is it worth talking about quality in general, if we allow that one of the indicators can be zero? The evaluation of quality is the subject of the study of science, which is called qualimetry The one of founders of qualimetry was G. Azgaldov. However, there are still a lot of people willing to write about the evaluation of quality so that not a word to say about qualimetry. They did not read the books of Azgaldov, and some even dispense with the term qualimetry. Without knowing the qualimetry, we make an obvious methodological error in the assessment of quality. What can be done with digital indicators "from zero to ten"? It's obvious: fold it! You can also divide the amount obtained by the number of indicators: you get a certain "average"... with "zero" quality for one indicator. The quality of the "individual indicator" does not compensate for the lack of quality in another indicator. Too much quality in one does not cover shortcomings in the other: quality in quantitative terms is not summarized. Each indicator should be expressed in the form of a decimal fraction, with the norm (the standard of quality) being the score of 1. The cumulative quality of the object is expressed by multiplying the indicators. There can not be at least one indicator equal to zero (multiplying any integer by zero yields zero). Even if at least one indicator from many equals 0.1 or 0.2, the cumulative index falls so low that it makes us ashamed to talk about any quality! Another interesting and very important regularity for us is connected with the number of parameters for which an estimate is made - as the number of parameters increases, the quality assurance problems increase dramatically. Thus, a conditionally "permissible" cumulative estimate of 0.7 with a number of indicators equal to 7 can be obtained only if each of them is not lower than 0.9. It is necessary to think whether it is worthwhile to express an abstract "score" on the evaluation of the quality of work, which in one case "pulls" for two thousand hours, and in another characterizes the work of the library for many years?
