SciELO - Scientific Electronic Library Online

 
vol.33 número1Innovation for inclusive development: mapping and auditing the use of ICTs in the South African primary education systemImplementation of machine learning techniques for prognostics for railway wheel flange wear índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados

Artigo

Indicadores

Links relacionados

  • Em processo de indexaçãoCitado por Google
  • Em processo de indexaçãoSimilares em Google

Compartilhar


South African Journal of Industrial Engineering

versão On-line ISSN 2224-7890
versão impressa ISSN 1012-277X

S. Afr. J. Ind. Eng. vol.33 no.1 Pretoria Mai. 2022

http://dx.doi.org/10.7166/33-1-2569 

GENERAL ARTICLES

 

Exploring the disconnect between the bodies of literature pertaining to socio-technical transitions and technology management (part 2): a linkage analysis

 

 

I.H. De KockI, *; A.C. BrentII

IDepartment of Industrial Engineering, Stellenbosch University, South Africa
IISchool of Engineering and Computer Science, Victoria University of Wellington, New Zealand

 

 


ABSTRACT

The importance and value of integrating the concepts of technology management with that in the socio-technical transitions literature has been highlighted in the literature. However, a disconnect still exists between these two bodies of literature. Therefore, this series of two papers investigates this disconnect from two perspectives. Part 1 investigates the disconnect by means of a bibliometric analysis that highlights the limited overlap and integration between technology management and socio-technical transitions. This paper, Part 2, enriches the investigation with a systematic and in-depth exploration of the literature bases used by the respective bodies of literature to gain additional insights into the level of integration (or the lack of it) between the socio-technical transitions literature and that on technology management. Similar to Part 1, this paper also finds that, even though these two fields are not integrated from a conceptual or theoretical perspective, it is evident that, to some extent, they share scholarly roots.


OPSOMMING

Die belangrikheid en waarde van die integrasie van die konsepte van tegnologiebestuur met dié in die sosio-tegniese oorgangsliteratuur is in die literatuur uitgelig. Daar bestaan egter steeds 'n skeiding tussen hierdie twee literatuurliggame. Daarom ondersoek hierdie reeks van twee artikels hierdie ontkoppeling vanuit twee perspektiewe. Deel 1 ondersoek die ontkoppeling deur middel van 'n bibliometriese analise wat die beperkte oorvleueling en integrasie tussen tegnologiebestuur en sosio-tegniese oorgange uitlig. Hierdie referaat, Deel 2, verryk die ondersoek met 'n sistematiese en diepgaande verkenning van die literatuurbasisse wat deur die onderskeie literatuurliggame gebruik word om bykomende insigte te verkry oor die vlak van integrasie (of die gebrek daaraan) tussen die sosio-tegniese oorgange literatuur en dit oor tegnologiebestuur. Soortgelyk aan Deel 1, vind hierdie referaat ook dat, al is hierdie twee velde nie geïntegreer vanuit 'n konseptuele of teoretiese perspektief nie, dit duidelik is dat hulle tot 'n mate wetenskaplike wortels deel.


 

 

1 INTRODUCTION

Technology plays an undisputed role in the quest for sustainable development. The technology and innovation management literature provides many concepts that are central to understanding the role of technology in sustainable business development [1], and the importance of technology management in the context of sustainable development has been argued in the literature [2]. Recently, scholars have also argued the importance of integrating the concepts of technology management and socio-technical or sustainability transitions [3], [4]. However, a bibliometric analysis that compared the respective bodies of literature relating to technology management and socio-technical transitions found "no concrete evidence of integration or significant similarity in foundational concepts used in both bodies of literature" [5]1. De Kock and Brent [5] further proposed that the bibliometric analysis and subsequent findings be enriched by a systematic and in-depth assessment of the literature bases (i.e., the references used by the respective bodies of literature) to clarify further the level of integration and overlap between the technology management and socio-technical transitions literature, in order ultimately to provide a starting point for the development of an integration strategy between technology management and sustainability transitions.

The bibliometric analysis undertaken in Part 1 [5] of this two-part investigation considered 331 documents that resulted from a keyword search that was focused on socio-technical transitions, and 4 740 documents that resulted from a keyword search focused on technology management (see Table 1). It emerged that only two documents ([6], [7]) are present in both sets of documents. Therefore, in order to investigate further where the two bodies of literature overlap and possibly integrate, this paper explores the links between the socio-technical transitions and the technology management bodies of literature, based on the references used by each set of documents (given the data sets that were extracted, as shown in Table 1 and as described in detail in [5]), to elucidate the level of integration between these two bodies of literature. The literature bases (the references used by each article) of both document sets are compared, and this detailed comparison is used to identify overlaps in the literature that are used as the basis for the research in the bodies of literature, thus aiming to elucidate further the extent to which there is an overlap, and to what extent these two bodies of literature share intellectual roots.

 

2 MATERIALS AND METHODS

The methodology followed in this two-part investigation into the disconnect between technology management and socio-technical transitions is shown in Figure 1. As mentioned above, a bibliometric analysis was conducted in Part 1 that had two phases, while Part 2 deals with the linkage analysis (LA) and has five phases. Similar approaches to evaluating the landscape, overlap, and integration of bodies of literature have been used throughout the literature [8]-[11]. The remainder of this paper thus focuses on the LA, and the remainder of this section focuses on the LA methodology.

 

 

2.1 Linkage analysis (LA) method

For all the documents that were extracted and used in Part 1 [5], only two documents2 formed part of the combined search (see Table 1 ). Thus, in order to evaluate the level of overlap and integration between the two sets of literature, the references associated with both scientific networks were evaluated to identify references in documents in both the technology management and socio-technical transitions scientific networks. The linkage used in this research inquiry refers to the cross-network method that is applied to reveal the linkage between the two scientific networks.

Because only two documents were found in both data sets, the references used in the socio-technical transitions and technology management documents respectively were compared with the aim of identifying the references that are used in both scientific networks. Thus the two data sets exported from Scopus (containing 331 documents and a resulting 17 445 references in the socio-technical transitions network, and 4 740 documents and a resulting 112 498 references in the technology management network)3 were used in the LA.

The input data for the linkage analysis was exported from the Scopus website using the .txt output format. Each 'Entry' (i.e., a document resulting from the search) in the input file has a title, author list, and bibliography list. Each bibliography list contains multiple items, referred to here as 'References'. The comparison of the data sets was done in two separate exercises, each comparing the references found in one data set with the references found in the other data set with the aim of achieving two respective outputs:

i. A list of references from the technology management set of documents that are also present in the socio-technical transitions set of documents; and

ii. A list of references from the socio-technical transitions set of documents that are also present in the technology management set of documents.

Ultimately, the two lists referred to above were used in Section 3 to identify and highlight the level of integration and overlap between the two sets of documents. In order to perform the LA, the data sets extracted from Scopus were used in a process that had four steps (LA Phases 1 - 4), each of which is described below.

2.1.1 LA Phase 1: Data pre-processing

This phase primarily entailed the sanitation of the data sets (References (R)) in the bibliography list. For each reference in the respective data sets, the following operations were performed:

1. Normalisation of references (correction heuristic). After the normalisation process, all that remained was the title of the reference, and any additional (nonsense) text that was not removed by the heuristic. This included:

a. Conversion of all text (reference strings) to lowercase;

b. Converting unicode to ASCII4;

c. Replace all foreign glyphs with the nearest Roman equivalent, or remove them;

d. Remove all author and publication metadata;

e. Remove all common abbreviations;

f. Remove all punctuation and redundant white spaces; and

g. Remove all URLs, dates, and page numbers.

2. Combine all references (R) of an entry (E) with white spaces

3. Repeat processes in steps 1 and 2 for the second set of references.

The normalisation and combining of the references in both data sets (the data sets with the references from the two scientific networks) was followed by the similarity calculation phase. The aim was to determine how likely a combined Reference (R) (i.e., the output from steps 2 and 3 above) of an Entry i (i.e., R'E ) from the technology management set of references was to contain a single reference from the socio-technical transitions list of references, and vice versa.

The edit distance algorithm5was used during the similarity calculations. This algorithm found the best match for the references in the first set of references within the second set of references, and vice versa. The maximum value for the edit distance would be achieved if one had to insert a completely new reference into the reference list of the entry's references against which the reference was compared. Thus the value would be equal to the length of the reference string. Consequently, we calculated the similarity coefficient as being , where N is the length of the string and e is the edit distance of the string.

2.1.2 LA Phase 2: Similarity calculation

The similarity calculation phase had two steps. First, a core algorithm was applied to measure the likelihood that an Entry (E) would contain a bibliography item; thus reference (R), since the references could not be directly matched. The second step of LA Phase 2 was the higher-level operation that yielded the similarity value (v), which indicated how likely it was that an edit operation referred to each reference.

LA Phase 2a: Core algorithm

For the purpose of this study, a measure of how likely an Entry (E) contained a bibliography item, thus reference (R), was needed. Since the references could not be directly matched (owing to discrepancies in the format, spelling, etc. between references), a modified edit distance algorithm was used as the core algorithm. The edit distance algorithm yielded the number of edit operations (insertion, deletion, or substitution) necessary to ensure that E contained R. The maximum value of the edit distance was reached when the whole text of R had to be inserted into E. Hence the similarity value , where e is the number of edit operations.

LA Phase 2b: Higher-level operation

During this phase, the similarity value (v) was calculated. In order to calculate v:

Let E1 = the first set of entries;

Let E2 = the second set of entries;

Then each e e Ethad a set of references (bibliography entries), Re. Taking a higher-level view, the objective was then to know for each e e E1all q e E2that had an overlapping bibliography entry within them. Thus, in set notation, for each e e E1.

To determine this, each reference Rq was taken from some qeE2and compared with the full bibliography text of some e eE1. This yielded a similarity value (v) that indicated how likely it was that e had a reference to Rq.

2.1.3 LA Phase 3: Threshold filtering

This phase filtered out q's based on their similarity values, v's. If v was less than the threshold, then q was not included in Q.

To summarise, the process described in LA phases 1 - 3 compared all references (referred to as Rtm1 , Rtm2, Rtm3, and so forth in Figure 2) resulting from a document (referred to as 'TM Entry V in Figure 2) in the technology management scientific network with the references resulting from all the documents in the socio-technical transitions scientific network (referred to as RsttI, Rstt2, Rstt3, and so forth in Figure 2) to establish the similarity between each reference in the technology management network and each reference in the socio-technical transition scientific network, and vice versa. The key objective, as stated, was to determine which references were used and, for any that were used, the frequency of their use in both scientific networks. Owing to the significant inconsistencies found in the bibliographic data extracted from Scopus, a similarity calculation was used to determine which references were present in the scientific networks, since a direct comparison was not possible. Comparing the data sets 'as-is' would have yielded a far lower number of references, as a large number of references were not cited correctly and/or were not the same.

2.1.4LA Phase 4: Data analysis

The results from LA Phases 1 - 3 yielded more than 3.9 billion data entries - namely, the total number of similarity comparisons that were performed when comparing all references from the two scientific networks6. An 'entry' was a line item showing the document (Entry) from the first set of documents, the document (Entry) from the second set of entries (whose references were compared with those in the first set of entries), and the references in the second set that were similar to the references in the first set. Similarly, the output from the second set of entries was the document from the first set of entries whose references were, in turn, compared with those in the second set of entries, and the references in the first set that were similar to the references in the second set. Each reference (in both sets one and two) was given a unique identifier at the start of LA phase 1. This meant that references that were the same would have different unique identifiers. However, the duplication did not influence the final results, as care was taken not to include duplicated values. Nevertheless, the duplication of the references found in each set that evaluated the similarity between the references used in the two document sets was analysed, as this indicated the frequency of the reference in the scientific network of that specific Entry.

In Phase 4, the similarity results - namely, the percentage similarity between two references - were evaluated. With the vast inconsistencies between the referencing styles and the information included in the nearly 130 000 references that were used in this research inquiry, a 75% or greater similarity between two references was deemed to mean that two such references were the same reference. However, there were references with a 75% similarity that, upon further investigation, were found not to be the same reference. Because the primary aim of the research inquiry was to identify and evaluate the overlap and integration of these two bodies of knowledge, this did not have a significant impact on the results. Furthermore, in the final set of results (refer to Table A.1), each reference was checked against the raw reference data and corrected if required to ensure that the correct number of occurrences was reported.

2.1.5LA Phase 5: Results

The output from LA phases 1 - 4 was two data sets:

1. A data set containing all the TM references that were also present in the STT scientific network. Thus all the TM references that were shown in this data set had a minimum similarity of 75% with at least one STT reference. This data set also showed the number of times that each TM reference with a similarity score of at least 75% occurred in the STT scientific network. In addition, the frequency of the occurrence of the TM reference in the TM data set was also shown.

2. Similarly, a data set containing all the STT references that were also present in the TM scientific network. Thus all STT references that were shown in this data set had a minimum similarity of 75% with at least one TM reference. This data set also showed the number of times that each STT reference with a similarity score of at least 75% occurred in the TM scientific network. In addition, the frequency of the occurrence of the STT reference within the STT data set was also shown.

The output was subsequently analysed in order to identify the areas (based on the similarity between references used by both networks) where (significant) overlap(s) occurred. Three different kinds of overlap between the TM references and the STT references were considered:

1. The most prominent references in both data sets7 (i.e., in the data sets where an overlap had already been identified - the two data sets described above). This included:

a. the top 50% most prominent STT references that were also a TM reference; and

b. the top 50% most prominent TM references that were also an STT reference.

2. References with at least 10 instances / occurrences within both data sets.

3. References with at least 10 instances in one data set and five in the other:

a. a reference with an occurrence / instance of 10 in the TM data set and an occurrence / instance of five in the STT data set; and

b. a reference with an occurrence / instance of 10 in the STT data set and an occurrence / instance of five in the TM data set.

The three different sets of overlap set out above provided the titles of references that occurred in both the TM and the STT data sets, with a varying number of instances or frequencies that each reference occurred in each scientific network.

From the above results, which essentially entailed the articles that cited the same references - thus the articles from which references were present in both the TM and the STT data sets - a data set containing the articles that drew from the same theoretical foundations (in other words, used the same references) was compiled. This set of articles was subsequently evaluated, and is discussed in Section 3. This was also used to expand the set of articles that could be used to evaluate the overlap between the technology management and socio-technical transitions bodies of literature - in other words, to expand the set of the two identified documents in Table 1.

 

3 LINKAGE ANALYSIS (LA): RESULTS AND ANALYSIS

This section explores the linkages between the socio-technical transitions and technology management bodies of literature, based on the references used by each set of documents (given the data sets extracted, as described in Section 2), to elucidate the level of integration and overlap between these two bodies of literature, and to identify the extent to which these two bodies of literature share intellectual roots.

3.1 Linkage analysis results

The results of the various phases of the linkage analysis are outlined below. Table 2 and Table 3 show the references of the 4 740 technology management documents and the 331 socio-technical transitions documents respectively that had a similarity score of 75% or above (refer to Step 2: Linkage analysis in Section 2).

3.1.1 Data pre-processing outcome

The respective sets of references were normalised (see Section 2, LA Phase 1: Data pre-processing for the approach) in order to have two data sets that showed only the titles of the references. Each document, as well as each reference in each of the two sets, was given unique identifiers (within each set). Essentially, each data set contained the unique number for each entry, with the corresponding numbers for the references associated with each entry and the title of each reference. The titles were used in the similarity calculation.

3.1.2 Similarity calculation outcome

The outcome of the similarity calculation phase was a data set showing the similarity scores. Thus the similarity scores of all the references in the technology management set of documents were calculated, enabling the identification of all references associated with the technology management documents (Entries) that had a similarity score of 75% or more with the references associated with the socio-technical transitions set of documents. As mentioned in Section 2, references with a similarity score of 75% or higher were regarded as also being in the set with which they were compared. Similarly, the references associated with the socio-technical set of documents with a similarity score of at least 75%, and thus were also deemed to be present in the references associated with the technology management set of documents, were identified.

Table 2 and Table 3 summarise the results. Here it is important to note again that each reference in each document was given a unique identifier. Thus, if the same reference (R) was cited by a number of Entries ( E), the specific reference was counted in each instance where the similarity score was 75% or higher. From Table 1, 112 498 references were present in the technology management data set, and a large number of these references were cited by more than one of the technology management documents (Entries). This was not a matter for concern, as both the frequency of documents that overlapped and the content and/or focus of the references that overlapped were of interest here. This point is highlighted simply because the number of references that were found to be present in both data sets did not necessarily indicate the number of unique references (the identification of the unique references was dealt with separately).

Table 4 shows an example of the results yielded by considering which references in the technology management set of documents overlapped with references in the socio-technical transitions set of documents. The example in Table 4 shows that the article by Geels [12], "Technological transitions as evolutionary reconfiguration processes: A multi-level perspective and a case-study", was cited by five documents in the technology management set of documents, and by 122 documents in the socio-technical transitions set of documents. Similarly, the data on the overlap of references cited by the socio-technical set of documents that were also cited by documents in the technology management set were identified.

Subsequent to the data gathered from the results of the similarity analysis, a further analysis was done on the results (the two data sets described above, as well as in the first part of LA Phase 5), to retrieve a list of references that adhered to the criteria set out in the second part of LA Phase 5 - i.e., (i) the most prominent references in both data sets, (ii) references with at least 10 instances / occurrences in both data sets, and (iii) references with at least 10 occurrences in one data set and five in the other. The outcome of this analysis is shown in Table A.1. By applying the set of criteria outlined above, a set of 119 references was yielded. These 119 references appeared 3 557 times as references used by documents in the technology management set of documents, and 1 538 times as references in the socio-technical transitions set of documents.

Table 5 summarises the different overlaps that were considered, as well as the corresponding number of articles found in each overlap group, and the number of articles that overlapped between these groups (refer to the matrix shown on the right in Table 5).

3.2 Analysis of overlapping references

In order to identify the areas of integration between the socio-technical transitions and technology management bodies of knowledge, and thus to identify to what degree these two bodies of knowledge overlapped and integrated concepts, and to what extent the concepts of the two bodies of knowledge were included in the respective fields, two approaches were taken:

1. An holistic analysis of the resulting overlap from three perspectives:

a. The resulting overlap (the 119 references highlighted above) were analysed, and insights and inferences drawn;

b. A cluster analysis; and

c. A correspondence analysis of the results from the linkage analysis.

2. The most significant overlaps in the above-mentioned set of references, which represented the overlap between technology management and socio-technical transitions, were analysed in order to elucidate further the overlap and integration (or lack of it) between the technology management and socio-technical bodies of literature.

3.2.1 Holistic analysis of the resulting overlap

In this section, the 119 references identified through the linkage analysis as references that were cited by Entries in the technology management and socio-technical transitions bodies of literature were analysed.

3.2.1. 1 Overview of the resulting overlap

The references considered to represent the overlap between the bodies of literature of technology management and socio-technical transitions (as shown in Table A.1) represent 0.007% and 0.001% of the references found in the socio-technical transitions and technology management bodies of knowledge respectively. These are arguably, by any standard, a (very) small percentage of the references under consideration, and so are the second (quantitative) indication (the first being that only two articles were present in both bodies of literature) of the disconnect between technology management and socio-technical transitions.

When the references with the highest number of occurrences in the technology management set were considered (shown in Table 6), it was evident that innovation was a prominent topic. In addition, strategic management and the literature dealing with competitive advantage, economics, and technological change featured strongly. It should be noted that there was a strong focus on the level of analysis at firm or organisational level. It was evident that 'technology management' did not explicitly feature as a key focus here, but instead was implied through the focus areas; and socio-technical transitions or sustainability transitions were not in this group of key focus areas.

However, when the references with the highest number of occurrences in the socio-technical transitions set that were also present in the technology management set were considered (shown in Table 7), there was a strong presence of documents that focused on transitions to sustainability and/or socio-technical transitions - possibly indicating an area of integration between the two bodies of knowledge under consideration. Also, as mentioned earlier (and as shown in Table 5), seven references fell within both the 'most prominent STT' and the 'most prominent TM' references; the focus of this (very limited) number of references was equally split between economics, innovation, social theory, and the social studies of technology; and one article focused on research methodologies.

When specifically considering the Entries in the technology management body of literature that referred to the references that dealt with transitions to sustainability (i.e., the articles in the technology body of literature that referred to the Reference in the 'most prominent' overlap group), the 29 occurrences of references focusing on socio-technical transitions from a sustainability perspective or sustainability transitions (as a broad term) yielded a corresponding 17 Entries in the technology management set of documents. Thus there were 17 technology management Entries that referenced these most prominent 'STT References' that specifically focused on socio-technical and/or sustainability transitions. Given that the objective was not only to identify references that overlapped, but also to expand the set of two identified documents in Table 2, the two articles that referenced the 'most prominent STT references' were in fact the two articles that were present in both bodies of literature (referred to in Table 1) - namely, the work of Dolata [7] and of Wells and Lin [6], clearly highlighting again the very limited overlap between these two bodies of knowledge.

Further considering the Entries, the references that dealt with transitions to sustainability (i.e., the articles in the technology management body of literature that referred to the references in the 'most prominent' overlap group), and by considering the keywords used in these Entries, it was clear that the key concepts that were addressed were in line with the findings when the most frequently used keywords were analysed for both sets of Entries in Part 1 of this investigation [5]. Figure 3 shows the most prominent keywords and 'keyword groups' found in these Entries. From this, and as highlighted earlier, innovation, technology, and sustainability were areas in which the technology management and socio-technical transitions bodies of literature overlapped. However, 'socio-technical transitions' was also present as a keyword.

 

 

From the above analysis, it can be concluded that there was an overlap at a high level and in respect of broad concepts (such as innovation, technology, and sustainability); but even these overlaps were based on a very small part of the data sets gathered at the start of the linkage analysis. The extent to which one had to delve into the data sets to find arguably minuscule overlaps in concepts was vast, and such concepts were then only indicative of overlaps and the integration of concepts at a high level and in broad terms.

The next section considers the correspondence and cluster analyses of the 119 references considered to represent the overlap between the bodies of literature of technology management and socio-technical transitions (as shown in Table A.1 ).

3.2.1.2. Correspondence analysis and cluster analysis

For the correspondence analysis, the standardised residual8 for each of the references (T1-T119) was calculated and subsequently displayed on a plot showing the varying degrees of the strength of the prominence of the references to either the technology management domain or the socio-technical transitions domain. The plot is shown in Figure 4. The greater the negative standardised residual, the less prominent was the relationship with a domain, and the greater the positive standardised residual, the more prominent was a reference in a specific domain. This also meant that the closer the standardised residuals were to zero, the smaller the difference in the prominence of such references in the respective bodies of literature, as well as a relatively significant overlap. Figure 4 shows the references in Table A.1, given their prominence in either the technology management or the socio-technical body of literature, based on their unit variance9. For example, a reference with a higher occurrence in the technology management body of knowledge will be placed closer to the technology management coordinate value in Figure 410. The closer the coordinates of a references (T1 - T119) are to zero, the more equal the occurrence in both bodies of literature, since the standard residual of such references are close to zero.

Subsequent to the correspondence analysis, and based on the calculated standardised residuals, a cluster analysis was performed. A cluster analysis aims to group data objects or data points on the basis only of information found in the specific data that describe such data points and the relationship between data points. The goal of cluster analysis is to group data objects together in a cluster that are similar (or related) to one another. The greater the similarity between the data points in a specific group, and the larger the differences between different groups, the more distinct the clusters.

For the purposes of this study, the goal of the cluster analysis was to determine whether, from a statistical perspective, there were references that could be grouped together (clustered) in order to draw insights from such clusters about the overlap landscape between the technology management and socio-technical bodies of literature (depicted here as the TM or STT domains). Again, as mentioned, the references used by the Entries of these two bodies of knowledge were considered in order to evaluate the overlap, as only two Entries were found in the overlap (refer to Table 1 ).

The correspondence and cluster analyses provided an abstraction from the individual data points presented in Table A.1 to the clusters in which those data points resided. In this specific case, when data points (i.e., the references) were grouped in the same cluster, it meant that such references had similar standardised residuals, and therefore had a similar prominence in their respective bodies of literature.

Figure 5 shows the dendrogram that was developed on the basis of the references presented in Table A.1 and the correspondence analysis discussed above. From Figure 5, depending on the selected linkage data, a number of different sets of clusters can be identified. In Figure 5, as depicted by the red line, a linkage distance of 31.581 yields five clusters. The selection of a linkage distance to identify clusters is a subjective decision. When considering the dendrogram in Figure 5, one can see that there are either three (should a linkage distance of between 50 and 60 be taken) or five (should a linkage distance of between 20 and 50 be taken) distinct clusters. The alternatives to this would be nine or 12 clusters if a linkage distance of about 11 or 5 respectively were selected. It is argued that, given the results found with the selected linkage distance (31.581 ) and the five clusters that this yields, sufficient insight is given into the overlap landscape for the purposes of this study.

Figure 6 shows the cluster membership of the five clusters resulting from the dendrogram, clearly indicating the domain within which each cluster (and therefore the references contained in each cluster) is more prominent. Clusters 1, 2, and 5 (and thus also the references associated with these clusters) have a greater prominence in the socio-technical transitions body of literature than in the technology management body of literature (relative to the other clusters, a greater positive standardised residual for socio-technical transitions, and relative to the other clusters a greater negative standardised residual for technology management), with cluster 2 (with average standardised residuals of -0,58 and 0,88 for technology management and socio-technical transitions respectively) having the average standardised residuals closest to zero, indicating a relatively high degree of similarity in the prominence of the references in the respective bodies of literature, as well as a relatively significant overlap. Clusters 3 and 4 have a greater prominence in the technology management body of literature than in the socio-technical transitions body of literature - namely, positive standardised residuals for technology management, and negative standardised residuals for socio-technical transitions.

Figure 6 shows the clusters across the set of references shown in Table A.1. The five clusters identified through the correspondence and cluster analyses described above are discussed below.

Cluster 1

As discussed above, the references in Cluster 1 were strongly associated with the STT domain, and thus with the socio-technical transitions body of literature. In other words, these references, relative to the other references identified in this overlap between the technology management and socio-technical transitions bodies of literature, consider topics that are explicitly and directly related to conceptual framings of socio-technical transitions and/or sustainability transitions - namely, the governance of socio-technical transitions, transition to sustainability theory, and a typology of socio-technical transition pathways.

Cluster 5

Similar to Cluster 1, Cluster 5 also had a stronger association with the socio-technical transitions body of literature than with technology management. However, the references were less prominent in the socio-technical transitions literature than the references in Cluster 1, and were more prominent in the technology management body of literature than the references in Cluster 1. When the references in Cluster 5 were considered, even though they were still strongly associated with socio-technical transitions and transition pathways from a conceptual perspective, some more applied and/or case studies based on socio-technical transitions were present. Another theme that was evident here was the social aspects that have to be considered when technology and the impact of technology are considered.

Cluster 2

Considering the outcome of the correspondence and cluster analyses, Cluster 2 had a slightly stronger association with the socio-technical transitions body of literature than with the technology management body of literature (see Figure 6). As might be expected, the references in this cluster were not concerned with socio-technical transitions, from neither a conceptual nor a practical/applied perspective - as was the case with Clusters 1 and 5; however, the references in this cluster were concerned most prominently with economics, technology-related topics (competing technologies, technological paradigms, and technology in organisations), innovation, and also articles dealing with research methodology. One article dealt with social science. Interestingly, the Brundland Report (1987) was also found in Cluster 2.

Cluster 4

The references in Cluster 4 had a stronger association with the technology management body of literature than with the socio-technical transitions body of literature (see Figure 6). When the 47 references in this cluster were briefly considered, themes and topics that were evident included technology-related themes (such as technology acceptance, technology roadmaps, and technical change), innovation, economics, organisational theory, strategic management and competitive advantage, and social science. A number of references were concerned with research methodology. However, here there was no clear coherence between the topics addressed by these references, like the references in Cluster 1 and Cluster 5, for example (but to a lesser extent). Both Cluster 2 and Cluster 4 had standardised residuals that were relatively close to zero for both the technology management and socio-technical transitions bodies of literature; thus the fact that no clear themes emerged from these clusters is not surprising.

Cluster 3

Cluster 3 is the one that had the strongest association with technology management, but not as strong as that of Clusters 1 and 5 with socio-technical transitions. Even though the association with technology management was relatively similar to that of Cluster 4, there was a noticeable difference in the prominence of these references with the socio-technical transitions body of literature compared with that of Cluster 4 (see Figure 6). The topics covered seemed slightly more defined in Cluster 3 than in Clusters 4 and 2, with innovation being a theme/topic that emerged quite strongly. Similar to Cluster 4, topics related to strategic management and competitive advantage and organisational theory were also present. A number of references addressed research methodology. It is worth noting that the references in this cluster were more obviously focused on the organisational level as the unit of analysis than clusters with a stronger association with the socio-technical transitions body of literature.

3.2.2 Most significant overlaps (absolute values)

When the 'most significant' overlaps of those identified in Section 2 under LA Phase 5: Results were considered - this was taken as all overlaps where the number of times that a reference was used / occurred in the technology management set and the number of times that a reference was used / occurred in the socio-technical transitions set was at least five11 or higher - 37 references emerged (as shown in Figure 7). Not surprisingly, of these 37 references, 30 were also in either the 'most prominent STT references' (that also had a presence in the technology management body of literature), or in the most prominent technology management references (that also had a presence in the socio-technical transitions body of literature), or in both. Also, as could be expected, given the outcome of the cluster and correspondence analyses above, all 15 references that were in Cluster 2 were also present in this set of references. Furthermore, if the references that formed part of Clusters 1, 3, 4, and 5 were considered, the distribution in terms of prominence between technology management and socio-technical transitions was relatively equal, with 10 references (Clusters 1 and 5 in Figure 7) being more prominent in the socio-technical transitions body of knowledge, and 12 references (Clusters 3 and 4 in Figure 7) being more prominent in the technology management body of knowledge.

As also could be expected - especially given the insignificant overlap between the two concerned bodies of knowledge when considering the small number of documents (two) found in the combined search (refer to Table 1) - the overlap considered in Figure 7 was not indicative of any specific dimensions across which these bodies of knowledge shared intellectual roots; some of these references were sources that discussed research methodologies and/or seminal papers, and so were expected to be present in these (and other trans- or multidisciplinary) fields. This was not because the content related to either socio-technical transitions or technology management, but rather because of the foundational concepts discussed in such documents, and they could be considered to have a high likelihood of being present in most multi- and trans-disciplinary bodies of knowledge that consider management sciences, engineering, technology, and social sciences. However, these 37 documents were analysed further to identify relevant overlaps and to infer - at least to an extent - the intellectual roots shared between the technology management and socio-technical transitions bodies of literature. The authors and year of publication of the references under consideration here are shown in Figure 8 and discussed below.

When considering the references shown in Figure 8, which were deemed the most significant overlaps in terms of the references found in the TM and STT bodies of knowledge respectively, eight were concerned with the science of research and/or research methodologies (denoted 'RM' in Figure 8); these were the studies of Strauss and Corbin (1998), Eisenhardt (1989), Yin (2002), Miles and Huberman (1994), Latour (1978), Glaser and Strauss (1967), Kuhn (1962), and Eisenhardt and Graenber (2007). These eight references could be considered indicative of an overlap in respect of research methodologies and/or approaches, but were not indicative of the dimensions across which these bodies of knowledge shared intellectual roots.

Of the remaining 29 references (thus excluding the references that were concerned with the science of research and/or research methodologies):

Seven focused on economics, economic development, or economic theory (denoted with 'E' in Figure 8). These were the studies of David (1985), Granovetter (1985), Dosi et al. (1988), Schumpeter (1942), Schumpeter (1961), Nelson and Winter (1982), and North (1990).

Seven focused on innovation and innovation studies (denoted with 'I' in Figure 8). These were the studies of Henderson and Clark (1990), Rogers (1995), Lundvall (1992), Christensen (1997), Geels (2004), Hekkert et al. (2007), and Van de Ven et al. (1999).

Ten focused on technological related themes such as technology adoption, technological change, social studies of technology, and technological development (denoted with 'T' in Figure 8). These were the studies of Artur (1989), Hughes (1983), Dosi (1982), Mackenzie and Wacjman (1999), Kemp et al. (1998), Bijker and Law (1992), Anderson and Tushman (1990), Geels (2002), Orlikowski (1992), and Bjiker et al. (1987).

Two focused on social studies (denoted with 'S' in Figure 8). These were the studies of Giddens (1984) and DiMaggio and Powell (1983). The Brundtland report (1987) (denoted with 'B' in Figure 8) was also part of this set of references.

Two documents - the studies of Bijker (1995) and Geels (2007) - focused on socio-technical change (denoted with 'STS' in Figure 8). However, Bjiker (1995) did not consider socio-technical change from a sustainability perspective, but rather described where technologies come from and how societies deal with them. The work by Geels (2007) considered various transition pathways for development along 'technological trajectories', but also not with a specific focus on sustainability.

It is interesting to note that two of the above-mentioned 29 articles - the studies of Hekkert et al. (2007) and Geels (2007) - formed part of the 'STT Entries' (i.e., the 311 socio-technical transition articles referred to in Table 1). None of the 29 articles under consideration here was also found in the 'TM Entries' (the 4 740 technology management articles referred to in Table 1).

 

4 DISCUSSION

The linkage analysis highlighted the most prominent areas of overlap between the technology management and socio-technical transitions bodies of literature, based on the references that the documents (Entries) in these bodies of literature cited. The linkage analysis yielded 119 references (out of a possible 17 445 socio-technical transitions references and 112 498 technology management references) that were present in both bodies of literature, thus representing the overlapping documents cited by the respective bodies of literature. As stated, significant or prominent overlaps were identified (refer to LA Phase 5: Results in Section 2); however it is argued that the criteria used to identify such significant or prominent overlaps were justified, given that they allowed for all overlaps of five or more to be included in the set, as well as any overlaps that were less than five but that were in the top half of references in either one of the bodies of literature. This meant that 0.007% of the socio-technical transitions references were also technology management references, and 0.001% of the technology management references were also socio-technical transitions references - the second quantitative indication that there is a disconnect between these two bodies of literature.

The areas of focus that emerged when the references with the highest number of occurrences in the respective bodies of literature, as well as the seven references that fell within both the 'most prominent STT' and the 'most prominent TM' references, were considered (refer to Table 6 and Table 7), were innovation, strategic management and competitive advantage, economics, technological change, socio-technical transitions, and social studies. An expanded set of articles was established, and the keyword analysis of this set of articles again highlighted that the areas of (limited) overlap were strongly geared towards innovation- and technology-related concepts. Interestingly, only here (at this significantly detailed level of analysis) did technology management and socio-technical transitions feature.

The correspondence and cluster analyses highlighted similar findings; the areas of overlap related to:

1. science of research and/or research methodologies;

2. economics, economic development, and economic theory;

3. innovation and innovation;

4. technology-related themes such as technology adoption, technological change, social studies of technology, and technological development;

5. strategic management and competitive advantage;

6. social studies; and

7. socio-technical change.

It is interesting to note the trend that, as one moved through the clusters from those with the most prominence in socio-technical transitions to those with a stronger prominence in technology management, it was clear how the references increasingly dealt with a unit and level of analysis that was at the level of organisations in Cluster 3, as opposed to at the macro-level of society or the economy in Clusters 1 and 5.

Taking a step back, and considering the total number of references found in the two data sets (17 445 and 112 498 for the socio-technical transitions and technology management data sets respectively; refer to Table 1 ), the overlap discussed above (of 29 references) was arguably negligible. Even though insights have been gained from considering the overlaps, they remain apparently insignificant.

 

5 CONCLUSION

From the various analyses performed and documented in this investigation (both Part 1 [5] and the current Part 2), one could conclude that the level of integration between the fields of technology management and socio-technical transitions is tiny. The overlaps that have been highlighted throughout this study, and that are summarised in Section 4, are primarily in respect of key concepts that are present in both bodies of literature, but arguably only at an aggregate level. There are no overlaps that emerge in respect of conceptual framings that are fundamental to either technology management or socio-technical transitions. It could be argued that the overlaps highlighted in these papers are partly as a result of the nature of the two bodies of literature, in that they are inter-, trans-, and multidisciplinary.

Ultimately, from the research and analysis conducted and discussed throughout this study, and the multiple perspectives from which the overlap of and integration between technology management and socio-technical transitions have been considered, it is concluded that the fields of technology management and socio-technical transitions have not been integrated at a conceptual or theoretical level.

An interesting observation is that there is one technology management concept that emerges, even though, in the broader scope of things, it could be considered still to be a limited emerging theme in both bodies of literature: technology roadmapping. This theme was found three times in Cluster 3 and twice in Cluster 4. Even though this is not indicative of a significant overlap of conceptual framings, it can be noted as an emerging area where technology management and socio-technical transitions might have been integrated.

Even though there is clear evidence that the fields of technology management and socio-technical transition are not integrated from a conceptual or theoretical perspective, it is evident that they do share intellectual roots across a number of dimensions - primarily those concepts related to innovation and technology. However, the unit and level of analysis at which these key dimensions were used in their respective fields largely differed: the unit and level of analysis was at the level of organisations in Cluster 3, as opposed to the macro-level of society or the economy in Clusters 1 and 5; and the unit and level of analysis was at the level of organisations in Cluster 3, as opposed to the macro-level of society or the economy in Clusters 1 and 5.

It can thus be concluded that the integration of socio-technical transitions approaches, concepts, frameworks, and aspects with those of technology management theories and practices, and vice versa, are not adequately addressed in the literature. Given the role of technology, and of its management, more research efforts are required across these bodies of knowledge to address the grand challenges [3], [4] and to enable a just transition to sustainability.

An important area for future research would be the further integration of technology management and socio-technical transitions concepts. For example, the various modes of interaction that have been defined between technologies could provide for further clarification of technology management considerations. It is commonly accepted that the mode of interaction between technologies can shift from one mode to another, and so it is suggested that specific technology management strategies be developed for each mode of interaction. It is also important to note that multi-mode interactions between technologies are possible, and that technologies can thus interact according to a number of interaction modes. Incorporating these theoretical notions into future research would add further value to the debate about how technology should be managed in the context of transitions.

Given the comprehensive investigation into the extent to which technology management and socio-technical transitions have been integrated, and the conclusion that there is no concrete evidence of integration of or significant similarity in the foundational concepts used in both bodies of literature, it is evident that there is a need for more informed, nuanced, and sophisticated theories, frameworks, models, tools, and techniques to support our understanding of how and where to integrate the concepts of technology management and socio-technical transitions. Similarly, future research needs to provide guidance on how socio-technical transitions and technology management research could advance in a way that addresses critical issues regarding epistemological tensions, problem identification and definition, the selection of system boundaries, the unit and level of analysis, and the role of technology management research in relation to socio-technical transitions research, and vice versa.

 

6 ACKNOWLEDGEMENT

Funding: This research was partly funded by the National Research Foundation (NRF) of South Africa, grant number 106962.

 

7 REFERENCES

[1] Wagner, M., Bachor, V. & Ngai, E. 2014. Engineering and technology management for sustainable business development: Introductory remarks on the role of technology and regulation. Journal of Engineering Technology Management, 34, pp. 1-8.         [ Links ]

[2] Brent, A.C. & Pretorius, M.W. 2008. Sustainable development: A conceptual framework for the technology management field of knowledge and a departure for further research. South African Journal of Industrial Engineering, 19(1), pp. 31 -52.         [ Links ]

[3] De Kock, I.H. & Brent, A.C. 2017. New insights into socio-technical transitions: A technology management perspective. In 2017 IEEE Technology and Engineering Management Society Conference (TEMSCON), pp. 329-334.         [ Links ]

[4] De Kock, I.H. & Brent, A.C. 2017. Technology management from a socio-technical transitions perspective. In International Association for the Management of Technology (IAMOT), pp. 966-986.         [ Links ]

[5] De Kock, I.H. & Brent, A.C. 2019. Exploring the disconnect between the bodies of literature pertaining to socio- technical transitions and technology management (Part 1): A bibliometric analysis. Submitted for review, 2019.         [ Links ]

[6] Wells, P. & Lin, X. 2015. Spontaneous emergence versus technology management in sustainable mobility transitions: Electric bicycles in China. Transportation Research Part A: Policy and Practice, 78, pp. 371 -383.         [ Links ]

[7] Dolata, U. 2013. The transformative capacity of new technologies, 1st edition. Routledge.         [ Links ]

[8] Klavans, R. & Boyack, K.W. 2006. Identifying a better measure of relatedness for mapping science. Journal of the American Society for Information Science and Technology, 57(2), pp. 251 -263.         [ Links ]

[9] Sakata, I., Sasaki, H., Akiyama, M. & Sawatini, Y. 2013. Bibliometric analysis of service innovation research: Identifying knowledge domain and global network of knowledge. Technological Forecasting and Social Change, 80(6), pp. 1085-1093.         [ Links ]

[10] Ittipanuvat, V., Fujita, K., Sakata, I. & Kajikawa, Y. 2014. Finding linkage between technology and social issues: A literature based discovery approach. Journal of Engineering and Technology Management, 32, pp. 160-184.         [ Links ]

[11] Chappin, E.J.L. & Ligtvoet, A. 2014. Transition and transformation: A bibliometric analysis of two scientific networks researching socio-technical change. Renewable and Sustainable Energy Reviews, 30, pp. 715-723.         [ Links ]

[12] Geels, F.W. 2002. Technological transitions as evolutionary reconfiguration processes: A multi-level perspective and a case-study. Research Policy, 31(8-9), pp. 1257-1274.         [ Links ]

[13] Dolata, U. 2008. The transformative capacity of new technologies: How innovations affect sectoral change: Conceptual considerations. MPifG Discussion Paper No. 08/2. Cologne: Max Planck Institute for the Study of Societies.         [ Links ]

 

 

Submitted by authors 10 Aug 2021
Accepted for publication 03 Feb 2022
Available online 06 May 2022

 

 

ORCID® identifiers
I.H. De Kock 0000-0003-4136-7418
A.C. Brent 0000-0003-3769-4512
* Corresponding author imkedk@sun.ac.za
1 De Kock and Brent [5] is Part 1 of the two-part investigation into the disconnect between the bodies of literature pertaining to socio-technical transitions and technology management. As the title of this article states, this is Part 2.
2 The two documents that are present in both the technology management (TM) and the socio-technical transitions (STT) primary document sets are Spontaneous emergence versus technology management in sustainable mobility transitions: Electric bicycles in China [6], and The transformative capacity of new technologies [7].
3 It should be noted that, during the linkage analysis programming, each reference was given a unique identifier; thus, should two or more documents in either of the scientific networks cite the same document, this document would have a number of unique identifiers (equal to the number of documents in the primary document sets that cite that specific reference). However, this duplication was accounted for in LA phase 5.
4 ASCII (American Standard Code for Information Interchange) is a character encoding standard. ASCII codes represent text used in computers, telecommunications equipment, and other devices (http://www.asciitable.com/).
5 In computer science, 'edit distance' is a way of quantifying how dissimilar two strings (e.g., words) are from one another by counting the minimum number of operations required to transform one string into the other (Skiena, S. [1997]. The algorithm design manual, Springer, New York).
6 In order to assess the similarity between the references in both datasets, each of the 17 445 STT references was compared with each of the 112 498 TM references; thus 17 445 x 112 498 comparisons resulted in 1 962 527 610 data entries. Similarly, the opposite comparison (i.e., 112 498 TM references compared with 17 445 STT references) produced the same number of data entries; thus there was a total of 3 925 055 220 similarity comparison data entries.
7 This refers to the data sets in which an overlap had already been established (i.e., the output datasets described above with an acceptable similarity score, as discussed in LA Step 4: Data analysis) between the two scientific data sets.
8 The standardised residual is calculated by dividing the residual (which is the difference between the observed and the predicted value of some variable) by the square root of the residual mean square. This produces scaled residuals that have, approximately, a unit variance.
9 Variance is a measure of variability, defined as the expected value of the square of the random variable around its mean (see also the previous footnote).
10 It should be noted that the coordinate value is used as the standardised residual alongside an arbitrary value (that is the same for all References) in order to show graphically the spread of References, and that some are more strongly linked to the technology management body of literature while some are more strongly linked to the socio-technical transitions body of literature.
11 The average overlap across the 119 references in Table A.1 was 4.3; thus the most significant overlaps were those that were above average - at least 5.

 

 

APPENDIX A

 


Table A.1 - Click to enlarge

 

APPENDIX B

 


Table B.1 - Click to enlarge

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons