Analysis of factors affecting information-seeking behaviour of the Ministry of Administrative and Bureaucratic Reform's policy analysts

A. INTRODUCTION Information-seeking behaviour is a human trait encompassing the cognitive process of recognising the need for, locating, and using information from various sources (Puri, 2013). It is a necessary part of filling gaps in one's knowledge (Olalekan et al., 2015). Hence, information-seeking behaviour arises due to an individual's demand for information, whether casual or professional. The State Civil Apparatus (ASN), particularly policy analysts, are professionals who heavily rely on information. These policy analysts need to stay updated to enhance their abilities and competencies, as well as strengthen their analytical prowess, enabling them to revise and develop policy in line with Regulation Number 45 from the Minister of Administrative and Bureaucratic Reform (Indonesia, 2013). This research investigates how job descriptions and capability specifications impact the information-seeking tendencies of policy analysts at the Ministry of Administrative and Bureaucratic Reform. In the policy development process, analysts need to navigate various stages, including agenda setting, policy creation, policy adoption, policy implementation, and policy evaluation (Alamsyah, 2016). They require methodological, technocratic, and political skills to analyse and employ information to gain political stakeholder support (Institution of State Administration, 2015). These skills also aid policy analysts in activities like research, policy analysis, policy recommendations, communications, coordination, advocacy, consultations, policy negotiations, and policy review publication (Indonesia, Kementerian PANRB, 2013). In the Ministry of Administrative and Bureaucratic Reform, 71.6% of the civil servants, totaling 391 individuals, were born between 1980 and 2000. Of these, 158 individuals or 28.9% are policy analysts, a substantial portion given that there are only 546 civil servants in the ministry. These policy analysts have job duties as described in the Ministry of Administrative and Bureaucratic Reform's Regulation Number 45 (Indonesia, Ministry of Administrative Reform and Bureaucratic Reform of the Republic of Indonesia, 2013). However, it remains unclear how their information-seeking behaviours are influenced by their job descriptions and capability specifications. This question is significant as policy analysts are integral to the formulation and implementation of bureaucratic reform policies, which inherently rely on information. Thus, this study aims to investigate two research questions: 1) Does the job description positively impact policy analysts' information-seeking behaviour in the Ministry of Administrative and Bureaucratic Reform? 2) Do capability specifications positively influence the information-seeking behaviour of these policy analysts? Accordingly, this study's goal is to determine if job descriptions and capability specifications affect the information-seeking behaviour of policy analysts within the Ministry of Administrative and Bureaucratic Reform. The hypotheses are: 1) Job descriptions have a positive impact on the information-seeking behaviour of these policy analysts; 2) Capability specifications have a positive effect on these analysts' information-seeking behaviour. Prior studies on information-seeking behaviour cover related topics like professional influences, driving factors, and information needs. This study brings novelty in its focus and subject matter. Moreover, the research's findings may provide useful insights and recommendations for improving service ease, access, and fulfilling the requirements of ASNs, particularly policy analysts in government agencies. B. LITERATURE REVIEW 1. Information-Seeking Behaviour Information-seeking behaviour refers to the intentional actions of an individual who recognizes a need for information, then finds and uses specific information from certain resources. Multiple models for understanding this behaviour exist, proposed by theorists such as Wilson, Johnson, Krikelas, Marchionini, and Leckie (Kundu, 2017). The model adopted in this research is that of Leckie et al. This model comprises several components connected to information-seeking behaviour. Firstly, the model considers functions and job responsibilities, signifying that every role has unique duties and functions that influence the type of information that needs to be sought and used (Kundu, 2017). Secondly, it accounts for the characteristics of information needs, which relate to demographic factors, context, frequency, predictability, the importance level of the problem within information needs, and its complexity. The third component looks into factors influencing the information search, such as the types of information sources (informal and formal), and the understanding of these sources (Kundu, 2017). Lastly, the model assesses the final outcome, which is determining whether the discovered information is suitable or not for the user's requirements (Kundu, 2017). 2. Policy Analyst A policy analyst is a functional role within the State Civil Apparatus (ASN) entrusted with conducting policy studies and analyses. They apply principles of professionalism, accountability, integrity, efficiency, and effectiveness to accomplish specific objectives or solve public issues (Regulation of the Ministry of Administrative and Bureaucratic Reform Number 45, 2013). According to Article 3 section 2 of the same regulation, a policy analyst is a career role, comprising of initial, junior, middle, and senior levels. ASN mandates policy analysts to have at least a bachelor's degree in any relevant field and must have completed policy analyst training (Indonesia, National Institute of Public Administration, 2015). 3. Job Descriptions for Policy Analysts The core responsibilities of a policy analyst involve conducting policy studies and analyses. This includes policy research and analysis, providing policy recommendations, engaging in communication, coordination, advocacy, policy consultation and negotiation, and publishing results of policy reviews (Regulation of the Ministry of Administrative and Bureaucratic Reform Number 45, 2013). 4. Capability Specifications for Policy Analysts In addition to their responsibilities in policy analysis and review, policy analysts also contribute to the realisation of good governance. They aim to establish a bureaucracy that is free of corruption, collusion, and nepotism and is committed to serving the public (Regulation of the Ministry of Administrative and Bureaucratic Reform Number 10, 2019). To inculcate these values within an organization, policy analysts are assigned to work units that match their expertise, allowing for more specialized and reliable policy creation. 5. Prior Research Numerous studies have explored the information-seeking behaviours of various professionals. A study by Guclu and Can (2015) indicated that the information-seeking behaviour of police officers in Turkey was influenced by the knowledge acquired from their experiences, interactions with colleagues, and work-related documents. Hence, their tenure and role had a significant impact on their information-seeking behaviours. However, the study didn't thoroughly elaborate on the specific duties of the police officers. Dasgupta et al. (2017) observed that the information-seeking behaviour of paramedic professionals in India was influenced by the latest scientific advancements in their field, their professional duties and functions, and immediate problemsolving requirements regarding medical cases. Nevertheless, their study didn't fully delineate for which activities the information was required. Research conducted by Kurnia (2014) and Puspita (2014) suggested that the information-seeking behaviours of professionals, specifically teachers and preachers, were driven by their job functions and roles, the characteristics of their information needs, their use of information sources related to their professions, and the relevance of the information they found to their work. Although these studies described behaviours related to information-seeking in professionals, including factors like fulfilling their information needs and elements related to their professions, they didn't elaborate on competency standards.

This research was carried out employing a quantitative approach through the method of surveying. The study focused on 158 policy analysts aged between 22 and 42 years, from entry to middle level. A systematic random sampling technique was employed to select the sample, and by using the Slovin formula (Sugiyono, 2019), it was determined that at least 113 respondents were needed. An online questionnaire was sent to 158 policy analysts via email and messages in the WhatsApp group titled “Analis Kebijakan.” The data collection took place from March 23 to April 8, 2022. In total, 171 responses were collected as some respondents had to attempt the questionnaire multiple times, and out of these, 118 were found to be processable, which exceeded the minimum requirement of 113 responses with 5% margin of error, and 95% confidence interval. Therefore, these 118 responses were analysed to get a more comprehensive representation of the data (Sugiyono, 2019). The variables in this research were divided into job descriptions (X1), capability specifications (X2), and information-seeking behaviour (Y), with a total of 64 indicators included in the questionnaire. Data collection occurred from March 23 to April 8, 2022, using Survey Monkey. The questionnaires were distributed via email and through the “Analis Kebijakan” WhatsApp group and were completed by the respondents themselves. The data analysis and processing utilised SPSS version 26 for nominal data (like gender, age, position, civil servant class, and tenure) and Partial Least Square-Structural Equation Modeling (PLS-SEM) via SmartPLS 3 for analysing the outer models, inner models, and hypotheses (Ken Kwong, 2013; Hamid & Anwar, 2019; Hair, et al., 2021).

1. Profile of the Ministry of Administrative and Bureaucratic Reform of the Republic of Indonesia The Ministry of Administrative and Bureaucratic Reform of the Republic of Indonesia serves the role of assisting the Indonesian President in areas relating to the development of state apparatus and bureaucratic reforms (Indonesia, 2015). The Ministry's vision involves fostering professionalism and integrity with an end goal of achieving a government of international standards, which aligns with the President and Vice President's vision of “Indonesia, a sovereign developed country, self-reliant, and based on mutual cooperation” (Indonesia, Ministry of Administrative and Bureaucratic Reform, 2020) In order to actualise this vision, the Ministry has set out several missions: first, to develop bureaucratic institutions and governance that are slim, agile, integrated, and electronically based (digital bureaucracy). Second, to cultivate personnel who are adaptive, professional, competitive, and globally conscious. Third, to evolve a transparent and accountable system of managing the performance of government agencies. Fourth, to foster a professional, independent, and integrity-driven supervisory system. Fifth, to realise public services that are clean, accountable, and beneficial. Lastly, to enhance the quality of bureaucratic reform management. In line with the Regulation of the Ministry of Administrative and Bureaucratic Reform Number 9 Pertaining to the Code of Ethics and Conduct for Employees of the Ministry of Administrative and Bureaucratic Reform (2017), the Ministry upholds certain values such as integrity, professionalism, and accountability. Integrity entails fulfilling duties honestly and consistently, exemplifying praiseworthy behaviour, and exhibiting discipline and dedication based on norms and ethics. Professionalism involves accomplishing tasks effectively and comprehensively, in alignment with one's skills and capabilities, and delivering exceptional results through innovative collaboration. Accountability, the final ethic, signifies taking responsibility for all actions, behaviours, and duties in terms of both processes and outcomes (Regulation of the Ministry of Administrative and Bureaucratic Reform Number 9 Concerning the Code of Ethics and Code of Conduct for Employees of the Ministry of Administrative and Bureaucratic Reform, 2017). 2. Demography of Respondents Among the 118 survey participants, a majority were female, making up 57.6% (68 individuals) of the respondents, with males representing the remaining 42.4% (50 individuals). The age group that was most represented was 26 and 27 year-olds, both constituting 12.5% (15 individuals each) of the sample. Regarding education, most respondents held a bachelor's degree, accounting for 65.3% (77 individuals) of the sample. When considering work level, 72.9% (86 individuals) of the respondents were first-class junior level policy analysts. The respondents largely belonged to the deputy for human resources of the apparatus work unit (38.1%; 45 individuals). The predominant civil servant class among the respondents was junior superintendents (III/a), comprising 59.3% (70 individuals) of the sample. The majority of respondents had held their current position for a duration of 1 to 5 years (76.3%; 90 individuals). 3. Descriptive Analysis of Research Variables: Job Descriptions (X1) The role of the policy analyst, as detailed by the Regulation of the Ministry of Administrative and Bureaucratic Reform Number 45 (Indonesia, Kementerian PANRB, 2013), is represented by seven indicators as depicted in Table 2. Examination of the table indicates that policy analysts perceive themselves to be knowledgeable and proficient in their job duties related to public policy preparation. 4. Capability Specifications (X2) These specifications pertain to the policy analysts' areas of expertise as determined by the Regulation of the Minister of Administrative and Bureaucratic Reform Number 60 concerning the organization's structure (Indonesia, Ministry of Administrative and Bureaucratic Reform, 2021). This variable is marked by five indicators displayed in Table 3. The table reveals that most policy analysts are competent in performing their roles in public policy formulation within their respective specializations or work units. Additional indicators requiring multiple-choice responses are illustrated in Table 4. Table 4 demonstrates that, for the indicator “frequency of similar public policy formulation each month,” the predominant response was “rarely (once/twice per month)” (45.8%; 54 individuals). Concerning the indicator “factors influencing your information seeking in public policy formulation,” most respondents (82.2%; 97 individuals) cited “the level of complexity of certain public policy.” For the indicator “the level of complexity of the public policy you are preparing,” the most frequent response was “complex” (55.9%; 66 individuals). 5. Information-Seeking Behaviour (Y) The model used to investigate this behaviour is adapted from Leckie et al. (Kundu, 2017), emphasizing the outcomes of professional information seeking. This behaviour is represented by the dependent variable (Y) as outlined in Table 5. The table suggests that the participants' informationseeking behaviour is driven by their roles and responsibilities as policy analysts, and the information obtained is beneficial for their duties. 6. Research Model Confirmation The data in this study, derived from 118 participants' responses, is processed using PLSSEM to establish the research model's validity and reliability. The ensuing sections detail the PLS-SEM analysis phases. 7. Outer Model Evaluation The validity assessment involves convergent and discriminant validity. Convergent validity is gauged using the loading factor value (> 0.7 is ideal, Hair et al., 2019) and the Average Variance Extracted (AVE) (> 0.5 is ideal, Hair et al., 2019). Discriminant validity is assessed by each variable's cross-loading value (ideal value is 0.7, Hair et al., 2019). Several indicators with outer loading values below the ideal have been identified, including UT8–UT12, KA1–KA11, KA14–KA19, KA22, PPI1, and PPI4–PPI6. The AVE values of these indicators are comparatively lower than for other variables, hence these indicators are eliminated, leaving 37 indicators. Subsequently, the model is recalculated to generate a new outer loading value, as depicted in the path diagram. Figure 1 above shows that the value for each indicator is within the threshold (0.7) (Hair et al., 2019). Therefore, it can be said that the variable is accurate. At this point, 37 out of 64 indicators remain for further testing. Removed indicators are as follows: - UT8: I understand the publication preparation of public policy studies results. - UT9: I understand procedures for making public policy formulation training material modules. - UT10: I understand the procedures for making audiovisual public policy formulation training material modules. - UT11: I understand the procedures for developing a public policy guidebook. l UT12: I understand the procedures for composing public policy implementation instructions and technical instructions. l KA1: I have the ability to create policy on politic and law affairs. - KA2: I have the ability to create policy on government and regional autonomy affairs. l KA3: I have the ability to create policy on state administration affairs. - KA4: I have the ability to create policy on work culture affairs. - KA5: I have the ability to create policy on performance management and cooperation affairs. - KA6: I have the ability to create policy on formulation and coordination of policies for implementing bureaucratic reform. - KA7: I have the ability to create policy on formulation and coordination of policies on the implementation of accountability and supervision for state apparatus. - KA8: I have the ability to create policy on coordination of policy implementation and evaluation of bureaucratic reform, accountability, and supervision for state apparatus. - KA9: I have the ability to create policy on formulation of institutional system policies and management. - KA10: I have the ability to create policy on policy formulation and coordination of egovernance implementation. l KA11: I have the ability to create policy on institutions and governance of politics, law, security, and local government. - KA14: I have the ability to create policy on work culture reinforcement of the state apparatus. - KA15: I have the ability to create policy on talent management and capacity building for state apparatus. - KA16: I have the ability to create policy on performance improvement and reward system for state apparatus. - KA17: I have the ability to create policy on accelerating the digital transformation of civil state apparatus management. - KA18: I have the ability to create policy on formulation of public service policy systems and strategies. - KA19: I have the ability to create policy on digital public service transformation. l KA22: I have the ability to create policy on coordination and facilitation of public service best practice development strategies. - PPI1: The result of my information seeking is useful for policy problem formulation. - PPI4: The result of my information seeking is useful for policy recommendation. l PPI5: The result of my information seeking is useful for policy implementation. l PPI6: The result of my information seeking is useful for policy monitoring and evaluation. The Average Variance Extracted (AVE) values for the variables were analyzed: Job descriptions (X1, AVE = 0.6), Capability specifications (X2, AVE = 0.621), and Information-seeking behaviour (Y, AVE = 0.593). As the AVE values of the variables exceed 0.5, this indicates that the study possesses strong validity (Hair et al., 2019). The assessment was then extended to establish the cross-loading value of each indicator. All indicators exhibited cross loading values surpassing 0.7, fulfilling the discriminant validity threshold and proving superior when compared to other variables (Ghozali, 2014; Hair et al., 2019; Hamid & Anwar, 2019). Hence, the current indicators precisely measure and depict the variables in this study. 8. Reliability Test The reliability assessment, part of the outer model evaluation, is carried out to ensure the consistency of the research variables. This was conducted by observing the composite reliability and Cronbach's alpha value, which should ideally exceed 0.7 (Hair et al., 2019). For the variables observed, Job descriptions (X1, Composite Reliability = 0.913; Cronbach's Alpha = 0.888), Capability specifications (X2, Composite Reliability = 0.867; Cronbach's Alpha = 0.797), and Information-seeking behaviour (Y, Composite Reliability = 0.953; Cronbach's Alpha = 0.947), all showed composite reliability and Cronbach's alpha values above 0.7. This suggests that the indicators used in this study's variables are dependable. 9. Inner Model Evaluation The inner model evaluation in this study was carried out using PLS-SEM to examine the R-square value and identify potential influences among the variables (Hamid & Anwar, 2019). The study's R-square value was 0.462, or 46.2%, suggesting that the variable influences information-seeking behaviour by 46.2%, in line with the theory that R-square is utilized to evaluate the significant impact of specific independent variables on certain dependent variables (Anuraga et al., 2017; Ghozali, 2014). The remaining 53.8% could be attributed to variables external to the research model that were not examined, such as demographics, information awareness, or the information sources employed. 10. Hypothesis Test The hypothesis testing was conducted using the bootstrapping feature of SmartPLS 3, examining the R-square, t-statistics, and pvalue. For a hypothesis to be accepted, it should have a significance level below 0.05 or a t-value that exceeds the critical value (Hamid & Anwar, 2019), and a t-statistics value higher than 1.96 (Hamid & Anwar, 2019). The first hypothesis tested in this research, which posits that job descriptions positively impact the informationseeking behaviour of policy analysts in the Ministry of Administrative and Bureaucratic Reform, resulted in a t-statistics value of 8.477, exceeding 1.96, and a p-value of 0.000, less than 0.05. Consequently, the first hypothesis is confirmed. Conversely, the second hypothesis stating that capability specifications positively influence the information-seeking behaviour of policy analysts in the Ministry of Administrative and Bureaucratic Reform, revealed a t-statistics value of 1.817, which is less than 1.96, and a pvalue of 0.070, exceeding 0.05. Hence, the second hypothesis is disconfirmed. In summary, the first hypothesis test demonstrates a Tstatistics value higher than 1.96, thereby validating the first hypothesis, while the second hypothesis test yields a T-statistics value less than 1.96, leading to the rejection of the second hypothesis. 11. Predictive Relevance and Model Fit The determination of predictive relevance and model compatibility is crucial for the suitability of the PLS model (Hamid & Anwar, 2019). This stage uses the blindfolding feature in SmartPLS 3. The Q2 value serves as a reference point for the PLS model's predictive relevance. If the Q2 value exceeds 0, the model has predictive relevance. The Q2 value in this study was 0.261. Besides the predictive relevance value, the model fit can also be examined from the Normal Fit Index (NFI) value using the bootstrap feature of SmartPLS 3. The NFI value of this research model was 0.643 or 64.3%, signifying a good model fit capable of explaining empirical data. Additionally, this study found that the NFI value was higher than the R-square value, indicating a higher degree of dependability. 12. Main Findings This investigation delves into the impact of job descriptions and capability specifications on the information-seeking behaviour of policy analysts at the Ministry of Administrative and Bureaucratic Reform. The outcome of this study reveals a positive correlation between job descriptions and the information-seeking behaviour of policy analysts. However, capability specifications do not demonstrate a similar influence. When examining the taskrelated aspect, job descriptions exhibit the highest path coefficient value of 0.611. This suggests a positive effect of job descriptions on the information-seeking behaviour of policy analysts. On the other hand, capability specifications register a low path coefficient value of 0.140 during the role-related aspect analysis. This shows that capability specifications do not positively influence the information-seeking behaviour of policy analysts. In addition, the indicator “frequency of similar public policy formulation each month” received a majority of responses for “rarely or 12 times per month” (45.8%), suggesting that the public policies handled each month are rather diverse and dynamic. Regarding the indicator “factors influencing your information seeking in public policy formulation (you may choose more than one option)”, most respondents selected the “level of complexity of public policy” and “limited time to prepare public policy” options (82.2% and 62.7%, respectively). Therefore, the complexity of certain public policies is a significant factor influencing policy analysts' informationseeking behaviour. This is further corroborated by the last question, “the level of complexity of the public policy you are preparing,” which had 66 respondents (55.9%) answering “complex”. Additionally, given that policy formulation is time-consuming, policy analysts require timely and accurate information. 13. Practical Implications This study presents several practical implications. The first of these is that job descriptions notably influence the informationseeking behaviour of policy analysts at the Ministry of Administrative and Bureaucratic Reform. Job descriptions comprise elements from the Regulation of the Ministry of Administrative and Bureaucratic Reform Number 45 Pertaining to Functional Position of Policy Analyst and the Credit Score (2013). This includes responsibilities such as research and policy analysis, policy recommendations, communication, coordination, advocacy, consultation, policy negotiations, and dissemination of policy review findings. This endorses Leckie et al.'s model (as per Kundu, 2017) suggesting that work tasks catalyse the information-seeking behaviour of professionals. This research also provides insights for information suppliers, enabling them to understand the roles and responsibilities of policy analysts in the public policy formulation process, as well as the kinds of information required by these analysts. Consequently, it underlines the need for support from information management units like agency libraries that can manage the data and information produced by policy analysts from each department, allowing this data to be sustainably utilised in support of policy analyst duties, namely, public policy formulation. The second practical consideration is that this study's model can be applied in other organisations to examine the connection between job descriptions and the information-seeking behaviour of policy analysts or civil state officials in functional roles. 14. Theoretical Implications This study modifies the informationseeking behaviour model of Leckie et al. (as per Kundu, 2017). This model proposes that roles and job descriptions affect the outcomes professionals achieve when they seek information. The theoretical implications of this study point to the strong correlation between the Ministry of Administrative and Bureaucratic Reform's job description for policy analysts and these analysts' information-seeking behaviour. However, capability specifications do not appear to impact policy analysts' informationseeking behaviour. 15. Limitations Several limitations exist in this research. The first of these concerns the timeframe for gathering research data. As described in the methodology section, this survey was distributed over a 17-day period from March 23, 2022, to April 8, 2022. Future studies should allot more time for data collection to garner more responses. The second limitation involves the singular focus on one functional position, that of the policy analyst. Despite the fact that policy analysts form a large percentage of individuals in functional roles within the Ministry of Administrative and Bureaucratic Reform, in terms of both number and age, this organisation houses several other functional positions that are filled by civil state officers who fall within the same age range as the research subjects.

This study explores the impact of job descriptions and capability specifications on the information-seeking behavior of policy analysts within the Ministry of Administrative and Bureaucratic Reform. The findings reveal that job descriptions positively influence this behavior, thereby confirming their significant role in shaping the information-seeking actions of policy analysts. On the contrary, capability specifications don't seem to positively impact this behavior, indicating their lack of significant influence on the information-seeking behavior of policy analysts in this Ministry. This research contributes to the field of library and information science by extending studies that use Leckie et al.'s model to examine the information-seeking behaviors of state civil servants in Indonesian governmental agencies. It is anticipated that the Ministry of Administrative and Bureaucratic Reform will take note of these findings. By understanding the factors that significantly influence or fail to influence policy analysts, the Ministry can more effectively supervise and assess the competency and performance of these analysts. This can also enhance the efficiency and professionalism of policy analysts in future public policy formulation. Thus, the relevance of this study to governmental agencies is the suggested need for collaboration between different work units and libraries to manage public policy-related data and information, supporting the future work of policy analysts.