Developing the core principles of responsible knowledge management (rKM): Section 4.3 – Observations from the Google Scholar search

This article is Section 4.3 of Chapter 4 of a series featuring my Master’s thesis The Emerging Concept of Responsible Knowledge Management (rKM): Identifying and Formulating the Core Principles of rKM.

The second search process, in Google Scholar, required a markedly different approach compared to Scopus due to fundamental differences in how the two platforms operate. Where Scopus allows users to combine multiple search criteria into a single, highly specific query, resulting in a narrowed and refined set of results, Google Scholar does not respond to increased specificity in the same way. In fact, the more search criteria one enters into Google Scholar, the more the results tend to diffuse rather than narrow. This inverse behaviour required a change in strategy: rather than running one composite search string, the process was broken into several targeted queries focusing on distinct aspects of rKM that had produced fewer hits in the Scopus search.

Guided by the gaps identified in the Scopus output, five separate searches were conducted in Google Scholar that yielded manageable and thematically relevant results. The publication timeframe for all these searches was aligned with the Scopus query, limited to the period between 2020 and 2025:

1. “Responsible knowledge management” – 93 results
2. Sources citing Durst’s A Plea for Responsible and Inclusive Knowledge Management – 25 results
3. “Ethics in knowledge management” – 50 results
4. “Ethical knowledge management” – 35 results
5. “Knowledge management” AND “common good” within journals whose names contain Knowledge AND Management – 93 results

Other combinations were tested but proved either too broad or yielded no results. For example, the query “social responsibility in knowledge management” returned zero results, while “social responsibility” AND “knowledge management” produced over 24,000, far too many for practical screening. Similarly, “common good in knowledge management” gave zero results, and “knowledge management” AND “common good” resulted in over 4,800 hits, again too broad to process effectively. Queries such as “critical knowledge management” were problematic because the term “critical” was predominantly interpreted in the sense of “crucial” rather than “analytical”. Searches for “values-based knowledge management” produced no results, and “purpose-driven” AND “knowledge management” returned 1,350 results, another unmanageable amount.

To maintain a reasonable and manageable scale for a single-researcher study, the search efforts were limited to these five targeted queries. After removing duplicates already captured in the Scopus sample, a total of 240 unique text candidates were identified from Google Scholar. These were first screened based on title relevance, which yielded 94 potential sources. Further filtering through keywords, abstract and conclusions reading and access availability resulted in 27 items selected for full-text review. Following the close analysis, 13 items qualified into the final sample. Given the discretion allowed by integrative literature review process, one article (G27) was handpicked outside both Scopus and Google Scholar searches as it emerged as relevant, being a source of one of the other sample texts (G07) identified through Google Scholar. The handpicked article also meets the inclusion criteria applied to the rest of the sample.

Compared to Scopus, the Google Scholar sample was more diverse in both form and disciplinary origin, encompassing peer-reviewed journal articles, book chapters, and conference proceedings. The broader indexing parameters and less restrictive editorial curation of Google Scholar produced a more interdisciplinary selection of texts, requiring careful filtering through reading to ensure conceptual and methodological relevance.

The selected items span conceptual and empirical contributions to the emerging discourse on rKM, with recurrent themes including spirituality and phronesis, non-rational and aesthetic knowledge, organisational purpose, AI ethics, knowledge for the common good, and inclusive governance. Several of the included texts explicitly engage with future directions of KM, offering visionary or critical perspectives that align with the ethos of rKM. Others introduce philosophical or indigenous frameworks, contributing valuable normative depth to the sample.

Notably, the Google Scholar texts tended to be less concerned with managerial efficiency and more focused on ethical orientation, transformational knowledge practices, and epistemological pluralism. In several cases, KM was not the sole or primary focus but was situated within broader conversations about societal, spiritual, or educational transformation.

These characteristics contribute meaningfully to the grounded theory development, especially in highlighting other rationalities and values not prominent in mainstream KM discourse.

While the sample required close scrutiny for academic quality and relevance, many of the selected texts offered critical expansions or redefinitions of what counts as knowledge and how it should be managed responsibly. As such, the Google Scholar sample provides an essential complement to the more functionally oriented Scopus texts, contributing to the overall diversity, richness, and conceptual range of the final study corpus.

It is important to reiterate the methodological limitations associated with Google Scholar. Unlike Scopus, which provides transparent indexing criteria and field-specific filtering options, Google Scholar functions as a proprietary search engine with no public index or inclusion standards. Its content base is therefore a black box, and this applies not only to what is included but also to how results are ranked by relevance. The algorithm that determines search result order is not open to scrutiny or adjustment. Moreover, search results on Google Scholar can be inconsistent across devices or time periods. Identical search strings may yield different results depending on the machine, user location, or even the date of the search. This challenge in replicability and transparency introduces an element of unpredictability to the research process and encumbers the reproducibility of the search results. To mitigate this limitation, the search process was documented in detail, including a list of the exact search terms, time frame, and any additional filtering steps used (i.e. Boolean operators), so that another researcher could attempt to replicate the search. This transparency allows the reader to evaluate the adequacy of the dataset and the reliability of the results – and enables subsequent attempts to reproduce.

To complement the textual description of both searches, the sequential filtering steps are visualised in Figure 14 below. The diagram summarises how the initial sets of 82 Scopus and 296 Google Scholar results were progressively narrowed through screening, eligibility checks, and full-text analysis to arrive at the final rKM sample. The figure also shows where items were excluded along the way and makes transparent how the Scopus sub-sample (19 items) and the Google Scholar sub-sample (13 items, plus one handpicked) together formed the basis of the overall corpus.

To increase transparency and support the reproducibility of the sampling process, a visual summary of the inclusion and exclusion criteria applied across both Scopus and Google Scholar searches is also presented in Figure 15. Although the two platforms required different operational approaches, the final selection was guided by a shared conceptual orientation and consistent filtering logic, as illustrated below.

In addition to this visual overview, a few concrete examples of excluded texts are also provided. While the complete list of rejections is longer, Table 3 illustrates representative cases where the inclusion–exclusion criteria were applied in practice. These examples show that while many of the texts addressed related themes such as sustainability, responsibility, or inclusivity, they sometimes fell short either by treating these concepts in narrow or instrumental ways or by lacking engagement with KM as a theoretical or systemic field. Presenting these cases offers greater transparency into the selection process and clarifies the validity of the final sample.

To illustrate the link between the theoretical foundations of this thesis and the empirical analysis, Table 4 below presents a matrix (refitted from Isabella¹, p.14) that juxtaposes key critiques of traditional KM with verbatim excerpts from the rKM sample, and the abductive leaps made in their interpretation. The table illustrates the retrospective strand of the iterative analysis, showing how the sample was examined against the theoretical foundations of KM. This complemented the later, prospective strand, where texts were organised into categories and aggregate dimensions. By aligning the ‘preliminary organising categories’ drawn from the theory chapter with direct quotations from the empirical material, the table makes visible how ideas that arose from the rKM literature often stood in sharp contrast to the underlying assumptions of traditional KM.

Since rKM is an emerging concept, there is no established theoretical library on which to rely. The theoretical foundations chapter therefore necessarily reflected a bricolage of antecedent ideas. This bricolage served as a mirror against which the rKM sample was interpreted, highlighting where the empirical material contradicted, departed from, or reworked those older assumptions.

Whereas traditional KM was characterised in the theory chapter as fragmented, instrumentalised, economistic, and reductionist, the rKM sample articulated alternatives that were markedly different. The abductive leaps represented in the right-hand column show how these contrasts could be synthesised into more generalised understandings that ultimately fed into the three aggregate dimensions discussed later in this thesis.

The purpose of including this table is therefore twofold. First, it demonstrates how the theoretical chapter acted as an active sensitising frame against which the principles of rKM seemed to rail, its older orientations guiding the interpretation of the sample toward more promising threads. Second, it shows that the movement from traditional KM to rKM is not simply incremental but paradigmatic, with the empirical material repeatedly surfacing orientations and practices that ‘cure’ the deficiencies of earlier approaches while opening space for new, still-emerging trajectories of rKM. This table provides a view of how theoretical critique and empirical insights interacted abductively to generate the findings.

Next part: Section 4.4 – Distribution of the sample.

Article source: Koskinen, H. M. (2025). The Emerging Concept of Responsible Knowledge Management (rKM): Identifying and Formulating the Core Principles of rKM. (Master’s Thesis, LUT University).

Header image source: Created by Hanna M. Koskinen using ChatGPT.

Reference: