Investigation on the statistical methods in research studies of library and information science
| Published date | 06 November 2017 |
| Date | 06 November 2017 |
| DOI | https://doi.org/10.1108/EL-02-2016-0042 |
| Pages | 1070-1086 |
| Author | Jin Zhang,Yanyan Wang,Yuehua Zhao |
| Subject Matter | Information & knowledge management,Information & communications technology,Internet |
Investigation on the statistical
methods in research studies of
library and information science
Jin Zhang,Yanyan Wang and Yuehua Zhao
School of Information Studies, University of Wisconsin Milwaukee,
Milwaukee, Wisconsin, USA
Abstract
Purpose –The statistical method plays an extremely important role in quantitative research studies in
library and information science (LIS). The purpose of this paper is to investigate the status of statistical
methods used in the field, their applicationareas and the temporal change patterns during a recent 15-year
period.
Design/methodology/approach –The research papers in six major scholarly journals from 1999 to
2013 in LIS were examined. Factors including statistical methods, application areas and time period were
analyzedusing quantitative research methods includingcontent analysis and temporal analysis methods.
Findings –The research studies using statistical methods in LIS have increased steadily. Statistical
methods were more frequentlyused to solve problems in the information retrieval area than inother areas,
and inferential statistical methods were used more often than predictive statistical methods and other
statistical methods. Anomalyanalysis on statistical method uses was conducted and four types of anomaly
were specified.
Originality/value –The findings of this studycan help educators, graduates and researchers in thefield
of LIS better understand thepatterns and trends of the applications of statistical methods in this field, depict
an overall picture of quantitative research studies in LIS from the perspective of statistical methods and
discoverthe change patterns of statistical method applicationsin LIS between 1999and 2013.
Keywords Evaluation, Quantitative research methods, Statistical methodology, Temporal analysis
Paper type Research paper
Introduction
Library and information science (LIS) focuses on the creation, selection/control,
organization, retrieval, dissemination and use of information. Research in LIS has
transcended its traditional research boundaries into areas as rich, diverse and dynamic as
other sciences, such as computer science. More sophisticated research methods address
expanding and diverse research problems from the societal, contextual, human,
technological and theoretical aspects of information. Furthermore, the quantity of
information has grown dramatically. Researchers in LIS must deal with an enormous
number of Web pages on the internet, a cornucopiaof queries submitted to a search engine,
a gigantic data set in a portaltransaction log, massive posts and responses in a socialmedia
website, an astronomicalnumber of records in a database and large and various user groups
in a community depending on their specific research area. It is impossible to investigate
every research object in a target population because of the wealth of data, technical
constraints, tight research budgets, geographic limitations, time limits and more.
Fortunately, the characteristicsof the population can be inferred through observationsmade
in a representative sample of the populationusing proper statistical method (s).
EL
35,6
1070
Received15 February 2016
Revised11 April 2016
29April 2016
2 September2016
6 February2017
15February 2017
Accepted16 February 2017
TheElectronic Library
Vol.35 No. 6, 2017
pp. 1070-1086
© Emerald Publishing Limited
0264-0473
DOI 10.1108/EL-02-2016-0042
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0264-0473.htm
Statistical methods are crucial for research studies in LIS. There is a trend that more
statistical methods are applied to the field. However, systematical studies on the
evolution of statistical method applications in the field are hardly found in the literature.
Statistical methodology includes inferential statistics and predictive statistics. Inferential
statistics is the inductive process of drawing conclusions from samples that are subject to
randomvariation.Predictivestatisticsis the process of predicting values based on
existing data. The former concentrates on hypothesis testing, whereas the latter focuses
on correlation analysis and regression analysis. The overarching research problem of this
study is to examine the current state of the applications of statistical methodology in LIS.
The significance of the study is reflected as follows: it will assist both graduates and
educators in their quantitative research method education; help researchers in the field
understand the relationships between the statistical methods and their application areas;
depict an overall picture of quantitative research in LIS from the perspective of statistical
methods; and discover the change patterns and trends of statistical method applications in
LIS.
Literature review
Research in common parlance refers to a logical and systematic search for knowledge
(Kothari, 2004). The most widely used research methodologies can be classified into the
following categories: quantitative, qualitative or mixed research methodologies (Kothari,
2004). The use of statistics is one of the most useful and powerful tools in data analysis
(Vaughan, 2001). However, Hildreth and Aytac (2007) examined a random sample of 100
research articles selected from 23 LIS journals and showed that only a few studies in the
field used correlationalor inferential statistical analysis methods.
Previous literature has indicatedthe need for librarians and LIS professionals to involve
themselves in more research. Research productivity has been regarded as one of the most
important criteria to assess LIS faculty and schools (Meho and Spurgin, 2005). However,
Powell et al. (2002) revealedthat only 42 per cent of LIS practitioners in the USA and Canada
occasionally or frequently performed research related to their jobs. Turcios et al. (2014)
reported that among the investigated 1,880 articles published in LIS journals, only 16 per
cent qualified as scholarlyresearch papers.
Insufficient use of statistical methodology in research studies of LIS in the early years
has been found. From 1967 to 1976, a mere 10.3 per cent of 402 articles published in
College & Research Libraries (C&RL) used inferentialstatistics (Kim and Kim, 1979). A few
years later, Wallace (1985) investigated the 25 LIS journals with the highest impact factors
and found that only 6 per cent of the articles used inferential statistics. Enger et al. (1989)
examined the same 25 LIS journals and found that 11.1 per cent of articles used inferential
statistics. Compared with the findings of studies conducted in the 1970s and the 1980s,
Dilevko (2007) showed that the use of inferential statistics in journal articles read by
academic and public librarianshas increased between 2001 and 2005.
In the early twenty-first century, a few research studies provided insight into the
overview of the research methods applied in a wide range of LIS studies. Williams and
Winston (2003) examined 119 articles fromfive journals and emphasized the importance of
research methods, especially statistical methods, in academic libraries. Hider and Pymm
(2008) conducted content analysis to identify the distribution of overall research strategies,
data collection methods and type of data analysis in 20 high-profile LIS journals in 2005.
Similar to the findings of the previous studies,quantitative research methods occupied more
than half of the overall journalarticles (64.7 per cent), and the percentage was even higher in
the non-librarianship journals (69.1 per cent). Tuomaala et al. (2014) conducted a
Investigation
on the
statistical
methods
1071
To continue reading
Request your trial