Knowledge Management: Determine the Influencing Factors for Practicing at the Libraries in Bangladesh

Aim. The main objective of this research is to know the influencing factors for KM practices in the library field of Bangladesh from the user’s point of view. There - fore, the study examines the elements influencing Knowledge Management (KM) practice in Bangladeshi university libraries. Methods. Present research used a quantitative approach, by adopting printed survey questionnaire for data collection from the users of the University of Dhaka (DU) and the University of Rajshahi (RU) in Bangladesh. Statistical software IBM®-SPSS® was used for data analysis, and the “partial least squares” (PLS) method was used to test the proposed hypothesis. Results. The findings revealed that KM familiarity and perceptions in gender and current study level varied across users. This study found that all the proposed hypotheses are supported, i.e., the service quality and critical success factors are the most influencing factors for practicing KM in the DU and RU library.


Introduction
L ibraries are continuously facing challenges in adapting new technologies for information resources. Such technological improvements now force libraries to reconsider the programmes they provide (Musangi et al., 2019). Libraries are a focal point for knowledge and should provide resources for exam preparation, group projects, seminars, symposiums, etc. Various library resources are used to write a literature review for research purposes and aid students in the most desirable ways with the changing information. They should assist as open access for gathering reading materials about the latest technological advancements (James & John, 2018). Learning something new is one of the most precise human endeavours, and knowledge is its outcome. So, it means that knowledge has been the subject of human inquest from antique times, and what we get from the straight connection with the situation through our sensory system, which is dealt with by the brain, is called knowledge (Bolisani & Bratianu, 2018). KM, is an enterprise, as well as knowledge itself (Zimmer & Madeja, 2019), which in higher education has extended its reputation as a management concept with a lot to benefit (Odor, 2018). KM is a vital element and it widens the acquired knowledge by increasing the capability of institutions to be more innovative and setting them in a gainful position (Omotayo, 2015). KM could hold great potential for providing service value to its users for libraries in developing countries. Service value includes value development by designing and delivering better services, and improving institutional and consumer operational innovation (Islam et al., 2015a). As society is improved, librarians need to explain their current status and how to tackle the new knowledge society (Siddike & Munshi, 2012).
An emerging nation like Bangladesh has yet to recognize the concept of KM and benefits from such an effort. In Bangladesh, there are inadequate library personnel, insufficient library facilities, lack of financial support, absence of government and non-governmental organization patronage, and trained users that would play the proper roles in making libraries a central of KM initiatives (Siddike & Munshi, 2012). Therefore, this research is helpful to know the influencing factors for KM practices in the library field of Bangladesh from the user's point of view.
Using KM in LIS has also shown that it has helped access information services, created awareness for IT practitioners, improved the KS culture situation, and changed librarian work responsibilities. The study acknowledges that the major obstacles to KM use and implementation include a lack of KM awareness, the lack of experienced personnel, communication gaps, and KS's nonexistence culture. Sultana and Mostofa (2018), found that the working environment in NLB supports the implementation of KM but that some policy changes are required. Finally, the analysis revealed that NLB is an ideal location for KM implementation and offers numerous opportunities. The successful application of the KM system in Bangladeshi libraries is obstructed by many obstacles and is seriously impeded by institutional, organisational, and psychological barriers.

Influencing Factors of KM
Md Anwarul Islam et al. (2015b) stated that incorporating KM will contribute to the development and creativity in academic libraries, with new service outcomes. Providing a more comprehensive and intentional analysis of the critical success factors for implementing KM is essential. Organisations need to be conscious of the factors that will influence the efficiency of a KM initiative and be aware of them (Migdadi, 2009). KM achievement can be described as capturing and achieving the exact information to the accurate user and utilising this information to develop personal management. Thinking about the numerous perspectives on KM, the success of KM implies to the institutional staff how they can use the knowledge which leads to the organisational aids, i.e., decision-making process, enhancing production and business efficiency, and others (Muttaqi, 2020). In organisations, several problems influence KM (Mahmood et al., 2020). Similarly, Maria Koloniari et al. (2015) identified that KM strategy, culture, structures of the organisation, and human resources management are the most important critical success factors of educational libraries in Greek. Siddike and Islam (2011) identified nine critical factors, i.e., organisational culture, IT, KM process, administration, KM strategy, etc., which would assist in building a proper KM application in Bangladesh libraries.

Research Questions
The research was directed by the following research questions (RQs). The RQs were answered using the data collected from the questionnaire survey.

Methodology
This study was conducted in Bangladesh from October 2021 to December 2021 at the DU and the RU. Both universities continue to have a strong character in meeting the higher education needs of a considerable percentage of Bangladesh's population. This study adopted a quantitative approach with a survey questionnaire design. For the collection of quantitative data, a simple random sample was used. Each active user of the respective library in different categories had an equivalent possibility of being chosen as a part of the sample. 600 printed questionnaires were sent to library users of the DU and the RU in Bangladesh. The study found 499 valid responses (fully filled questionnaires by the users) for data analysis, with a response rate of 83.1%. The instrument employed for this purpose was a self-administered survey questionnaire and a 1-5 point Likert scale from 1 (strongly disagree) to 5 (strongly agree) and 1 (very low) to 5 (very high).
The first section of the questionnaire focused on demographic information, while the second section asked about research questions. A printed copy of the questionnaire was provided to the participants. A completed questionnaire was collected from the users by the nominated persons by the researchers with the library staff's help. Finally, the data were inserted into the statistical software IBM®SPSS® 20 for final analysis. Cronbach's alpha was used to determine the questionnaire items' reliability. The final alpha value was .94, indicating that the questionnaire was reliable. We also checked the internal consistency of the instrument. As shown in Table 3, Cronbach's alpha (α) of the variables varied from .84 to .93. Descriptive statistics were used to analyse the demographic information of the students. Mean, and SD were also derived from the users' responses. To see the differences among gender, and current study level in terms of KM familiarity and perceptions, non-parametric Mann-Whitney U and Kruskal-Wallis H tests were carried out, and a p-value of <.05 was considered significant. The Mann-Whitney U test is used to test whether two samples are likely to derive from the same population. The Kruskal-Wallis H test is a rank--based non-parametric test. It's an extension of the Mann-Whitney U test that allows comparing more than two groups. The hypotheses of the study were tested with SmartPLS 3. SmartPLS is software that uses the partial least squares (PLS) path modeling method to do variance-based structural equation modeling (SEM) (Hair et al., 2022). One of the most extensively utilised multivariate data analysis methods among business and social science experts is "partial least squares-structural equation modeling" (PLS-SEM), also known as PLS Path Modeling (Memon et al., 2021). Researchers can use SEM to evaluate a model's overall fit and to examine the structural model (Chin, 1998).

Research Conceptual Model and Hypothesis
A conceptual model was developed for DU and RU in Bangladesh to illustrate the relationship among the various factors for KM implementation in libraries. The proposed intuitive model is made up of five components. These components will assist in finding out which are the most influential factors for practising KM in libraries. Figure 1 displays the model that provides the basis for analysing the structural equation model (SEM). In the model, the quality of the library service, familiarity with KM, critical success factors, and challenges faced by the library are considered independent variables, and KM practice is regarded as the dependent variable.

Figure 1 Research Conceptual Model
Note. H -Hypothesis.
Source. Own research.
The research sought to test the four hypotheses based on the research questions, conceptual framework (see Figure 1), and the literature reviewed. Survey data tested the hypothesis to demarcate what factors are significantly associated with KM implementation in the library. Both path estimates and t-statistics were used to test the hypotheses. The PLS algorithm test was used to look at path coefficients, while the bootstrapping test was used to look at t-statistics. The bootstrapping test is a resample based on the existing data (Rijlaarsdam, 2007). The following hypotheses were tested; • Hypothesis 1: There is a significant relationship between the quality of the library service with KM practice. • Hypothesis 2: There is a significant relationship between KM familiarity issues with KM practice. • Hypothesis 3: There is a significant relationship between critical success factors with KM practice. • Hypothesis 4: There is a significant relationship between challenges faced by the library with KM Practice.

Variables, Coding, and Items of the Questionnaire
To ensure the validity of the research, the measurement items and variables were developed from prior studies. Some measurement items also were self-developed. The variables, coding, and items are listed in Table 1.

RKMLP1 -An important ingredient of KM is the expertise of LIS specialists in librarianship
RKMLP2 -Activities in a library's readers' service section, such as distribution of books, reference services, etc., are synonymous with sharing KM awareness RKMLP3 -KM helps in enhanced productivity or service quality Source. Own research.

Reliability Statistics
We checked the questionnaire's reliability using IBM®SPSS® Statistics. The questionnaire's reliability coefficient is shown in Table 2. The Cronbach's alpha test is 0.94, which is considered reliable because it is greater than .05.   Table 4 shows that male students are higher than female students.

Quality of the Library Services
The findings, presented in Table 5, noted the following results when the users were asked about the quality of the library services. The study found that most users agreed with the statement "Service of the library is very good" (M = 3.02; SD = 1.29). Where a significant number of the students agreed the "Membership process to this library is easy" (M = 2.86; SD = 1.42). Most users also agreed that "Staff is actively involved in better service of the library" (M = 3.05; SD = 1.24). Overall, it can be said that the majority of this question's mean score above three indicates a high degree of agreement among the users. Meanwhile, the value of SD obtained from the items noted that the findings were significant.

Familiarity with KM
The research noted the following results when the users were asked about the user's familiarity with KM. Present research findings indicated that the "Library needs to be conscious of critical success factors that would influence the implementation of KM" obtained the highest mean score (M = 2.76) with a SD of 1.17. Where "Relationship between KM familiarity issue and critical success factors" obtained the second highest mean score of (M = 2.55) with a SD of 1.09. "Relationship between KM familiarity issue and service value" obtained the third position with an overall average (M = 2.44) with a SD of 1.05. "Familiarity with KM" achieved the lowest mean score (M = 2.11) with a SD of 1.08. All these findings ranked on a 1-5 point Likert scale 1 (very low) to 5 (very high). The results showed that students of various departments have moderately low familiarity with KM (Table 6).

Critical Success Factors
The findings in Table 7 Table 8

KM Practices in the Library
The findings, presented in Table 9, show the frequency and percentage of the relevance of KM on library practice measured on a 1-5 point Likert scale 1 (strongly disagree) to 5 (strongly agree). The research distinguished the following results when the users were asked about the relevance of KM to library practice. The study identified "KM helps in enhanced service quality" as obtained the highest mean score (M = 3.36) with a SD of 1.17 for the relevance of KM on library practice, followed by the "Activities in a library's readers' service section" with the mean score of (M = 3.06) with a SD of 1.13. "An important ingredient of KM is the expertise of LIS specialists in librarianship" was considered less relevant by the users with an overall average (M = 2.76) with a SD of 1.29. Overall, it can be said that most of the mean values above three indicate a high rate of consent among the users. Meanwhile, the low value of SD indicated that the findings were considerable.

Mann-Whitney U Test for User's Gender and Personal Characteristics
The results of the "Mann-Whitney U Test" found statistically significant differences between gender and KM familiarity and KM perceptions for "How much familiarity with KM?" (Mann-Whitney U = 21193.00, p <.05). "Relationship between KM familiarity issue" (Mann-Whitney U = 21813.00, p <.05). Significant differences were not found between gender and KM familiarity and KM perceptions for "Relationship issue and critical success factors?" (Mann-Whitney U = 22668.50, p>.05), "Library conscious of critical success factor?" (Mann-Whitney U = 24615.50, p >.05) ( Table 10). Source. Own research.

The Kruskal Wallis H Test for the Current Study Level of Users with KM Familiarity and Perceptions
The result of the Kruskal Wallis test from Table 11 found that the statement "How much familiarity with KM" (p <.05),) and "Relationship between KM familiarity issue" (p <.05), and "Relationship issue and critical success factors" (p <.05) has a significant difference with the current study level. Table  11 also revealed that no significant difference was found with "Library conscious of critical factors that would influence the implementation of KM" (p >.05).  Table 12 shows a reliable and valid measurement model. All parameters were above the acceptable value of .6, indicating that the indications were reliable. The composite reliability and Cronbach's alpha values for the constructs were higher than the suggested value of .7, showing excellent internal consistency dependability. The constructs' AVEs were greater than the recommended value of .5, indicating that convergent validity was sufficient. The square root of the constructions' AVE values was greater than the correlations between the constructs, and all indicators loaded higher on their respective constructs, indicating appropriate discriminant validity.

Fornell & Larcker Criterion
Discriminant validity was also tested using the criterion suggested by Fornell and Larcker (1981). The results of both tests are reported in Table 13. Source. Own research.

Assessment of the Measurement Model
As shown in Figure

Assessment of the Structural Model
The purpose of assessing the structural model is to evaluate its validity and test the hypotheses. The constructs, also known as the latent variable, and the path that connects them make up the structural model, as shown in Figure 3. The path significance of the structural model is estimated by bootstrapping, a resampling technique. The bootstrap procedure produces t-values for each path in the model. tively, indicating modest, medium, and high effect sizes (Henseler et al., 2009), and The model's and parameter estimates' predictive relevance (Q²) are measured by how well they rebuilt observed values, which should be greater than zero (Chin, 2010). Examining the structural model revealed an adequate and valid model, as shown in Tables 14 and 15 and Figure 3. According to William W. Chin (1998), an R² value of .67 is considered significant, whereas values of .33 are medium, and values of .19 are weak. The R² values for RKMLP were large, demonstrating strong explanatory power. The dependent variables› predictive relevance (Q²) values were higher than the suggested value of zero, indicating that the model's predictive relevance was appropriate. A value less than .10 or .08 in SRMR and NFI values between 0 and 1 (Hu & Bentler, 1999) is considered a good fit. The study shows a good fit of the model (SRMR = .07; NFI = .78) because it is less than the recommended value of Linden T. Hu and Peter M. Bentler (1999). The effect size (f²) values were within the recommended values ranging from .001 to .119, demonstrating the independent variables' small and medium effect sizes.

Discussions and Major Findings
This paper sought to address three RQs. found statistically significant differences between gender and KM familiarity and KM perceptions for "How much familiarity with KM?" (Mann-Whitney U = 21193.00, p <.05). "Relationship between KM familiarity issue and service value" (Mann-Whitney U = 21813.00, p <.05). From the Kruskal Wallis test, the study also found that the statement "How much familiarity with KM" (P<.05), "Relationship between KM familiarity issue and service value" (P <.05), and "Relationship between KM familiarity issue and critical success factors" (P <.05) has a significant difference with the current study level. The third question was, what are the challenges with KM practices in DU and RU libraries in Bangladesh? From the findings of the study, it is indicated that the highest number of participants agreed with the statements such as "Lack of awareness" (M = 3.37; SD = 1.37), and "Losing information from employee's resignation and retirement" (M = 3.26; SD = 1.30) are the major challenges for KM practice at DU and RU libraries. At the same time, Islam et al. (2014) identified that lack of awareness is an important barrier to implementing KM. Similarly, Rajeev Verma and Kambalor Ramakrishna Jayasimha (2014) reported that employees have no time for KM, lack of understanding of KM and its benefits, lack of skill in KM techniques, organisation's processes are not designed for KM. They also stated that lack of funding for KM, and lack of commitment from top-level management are the challenges for KM implementation in the organisation. The path coefficients of the proposed model showed that all four hypotheses are supported at the .05 significant level.

Conclusion and Limitations
The study attempted to determine the influencing factors for KM practices in the DU and RU libraries in Bangladesh. In addition, this study tried to identify to what extent users' gender and study level impact KM familiarity. The study found that the quality of the library services, familiarity with KM, critical success factors, and challenges in implementing KM are the influencing factors for KM practices in Bangladesh's DU and RU libraries. The Mann-Whitney and the Kruskal Wallis tests found statistically significant differences between gender and current study level with KM familiarity and perceptions. The highest number of participants strongly agreed with the statements such as "Establishing a solid infrastructure for future development, and "Organisational ICT structure," which are considered the critical success factor. From the findings of the study, it is indicated that the "Lack of awareness," and "Losing information from employee's resignation and retirement" are the significant challenges for KM practice in DU and RU libraries. The hypothesis result showed that all the proposed hypotheses are supported and significant at the level of .05. This research provides a significant contribution because, comparatively, KM is new in the LIS context of Bangladesh and other developing countries (Islam, Sid-