Cognitive Mapping IJ.. - deseng.ryerson.ca - Ryerson University

6 downloads 13192 Views 512KB Size Report
article (e.g. in Word or Tex form) to their personal website or .... the literature and their domain knowledge of the situation, that the .... buy in to, the output.
This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/authorsrights

Author's personal copy

International Journal of Industrial Ergonomics 43 (2013) 304e313

Contents lists available at SciVerse ScienceDirect

International Journal of Industrial Ergonomics journal homepage: www.elsevier.com/locate/ergon

Cognitive mapping: Revealing the links between human factors and strategic goals in organizations Judy Village*, Filippo A. Salustri, W. Patrick Neumann Human Factors Engineering Lab, Department of Mechanical and Industrial Engineering, Ryerson University, 350 Victoria St., Toronto, Ontario M5B 2K3, Canada

a r t i c l e i n f o

a b s t r a c t

Article history: Received 15 April 2012 Received in revised form 16 January 2013 Accepted 7 May 2013 Available online

The authors propose cognitive mapping (CM), a tool used in operations and management research, as a way for Human Factors (HF) Engineers to understand the HF perspective of senior managers and others in manufacturing industries, and how HF aligns with strategic goals in the organization. This paper first presents a methodological review of various mapping methods. Options are summarized with respect to: how to elicit information; the role of the facilitator; mapping methods and analyses; and interpretation of the data. Second, we choose a mapping method and demonstrate its utility with a single participant. Results from the illustrative example show the visual nature of the tool in summarizing the perceptions of the participant. We suggest CM methods can help HF Engineers and others work with industry to identify actionable steps to integrate HF into daily practice in ways that support strategic organizational goals. Relevance to industry: : Aligning human factors to organizations’ corporate strategies will enhance its application, and therefore effectiveness. Such macroergonomic tools are needed to facilitate understanding by senior management of the strategic potential for human factors and to help create aligned HF initiatives. This paper presents a methodological review and illustrative example using cognitive mapping for this purpose. Ó 2013 Elsevier B.V. All rights reserved.

Keywords: Cognitive mapping Strategic goals Human factors Macroergonomics

1. Introduction and background 1.1. The need for a tool to link human factors to an organizations’ strategy Dul and Neumann (2009) suggest that human factors (HF) considerations would be more accepted and better internalized in organizations if they were understood by managers to contribute directly to the companies’ strategies. HF groups generally want to be proactively involved in design activities to prevent problems and they want to contribute to the organization’s goals. However, the challenge is that in their often limited “support” role, HF groups are frequently disconnected from management strategies and others affecting business such as engineering groups (Perrow, 1983; Dixon et al., 2009; Jensen, 2002). HF groups may be unaware of perceptions held by managers and engineers concerning HF, and how best to help achieve strategic goals. HF engineers (used synonymously here with Ergonomists) may be perceived as defenders of operators * Corresponding author. 2220 Badger Rd, North Vancouver, British Columbia V7G 1T1, Canada. Tel.: þ1 604 929 7243; fax: þ1 604 929 7280. E-mail addresses: [email protected], [email protected] (J. Village). 0169-8141/$ e see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ergon.2013.05.001

(Perrow, 1983), rather than being essential to achieving the organizations objectives, or integrated throughout the production systems design. While there is evidence that engineering changes are more difficult and costly when HF considerations are incorporated late in the design lifecycle (Miles and Swift, 1998; Seim and Broberg, 2010), HF engineers have difficulty accessing design groups. HF engineers are often consulted too late in the design process; they are seen as critical of the design (Broberg, 2007; Kirwan, 2000; Hendrick, 2008), thereby increasing costs and delays (Perrow, 1983). HF engineers can reduce the conflict described if they know the strategic goals of the design process and align HF with these goals. Waterson and Kolose (2010), in a large scale military defense organization, discussed some of the challenges experienced by the HF team with management perceptions of their function and purpose. Although the team had been in existence and had supported many functions for several years, the authors reported that it still lacked visibility and prominence in the overall organizational structure. It was referred to as a “bubble” on a flow diagram of an organization chart. It appeared that others in the organization were unaware of the value of involving the HF team.

Author's personal copy

J. Village et al. / International Journal of Industrial Ergonomics 43 (2013) 304e313

Integrating HF into an organization, like other business reengineering and quality efforts, is really an “organizational change” initiative because they affect multiple levels in an organization and require the interaction of different agents. The research on organizational change strongly suggests that the successful initiatives are the ones that are integrated into the “psychological” dimension, or mindset, of the organization (Zink et al., 2008). Similarly, to facilitate integration of HF among groups and functions throughout an organization, and especially to gain management commitment, HF must be aligned with the strategic objectives and business function strategies (Dul and Neumann, 2009; Drury, 2000; Genaidy et al., 2009; Neumann et al., 1999). The task for HF Engineers is the practical one of finding ways to gain management support and facilitate alignment of HF considerations with the strategic goals and business outcomes of the organization. This paper adapts “cognitive mapping,” from the operations management field, and provides an illustrative example of its use in HF. The goal of this methodological paper is to demonstrate a practical macroergonomics method for HF Engineers to facilitate alignment of HF considerations with strategic goals of senior management and engineers. 1.2. Objectives The objectives of this paper are to:  Introduce the cognitive mapping technique that can make apparent the perceptions about HF and how it relates to a company’s strategic goals;  Review methodological options for applying the cognitive mapping technique;  Choose one method of cognitive mapping for use by HF Engineers and demonstrate it with an illustrative example; and  Recommend how HF Engineers could use cognitive mapping collaboratively with senior management and other agents to improve discussion of and action on HF implementation. 2. Methodological review of cognitive mapping 2.1. What is cognitive mapping? A cognitive map is a graphical representation, or visual picture, of the content and structure of an individual’s belief system (Eden et al., 1992). The process of cognitive mapping was introduced into the management science field by Axelrod in 1976 (Markoczy and Goldberg, 1995). The basis of the theory originates in the psychological “personal construct theory” (Kelly, 1955). The personal construct theory posits that humans are scientists who are constantly trying to make sense of the world in order to act within and upon it. In trying to make sense of the world, people use a construct system, then compare any new information for similarities and differences and map these relationships to form their perceptions. The process is one of reflective comparison between currently held concepts and new information. The process of eliciting the map is most commonly performed using interview techniques and open-ended questions about a specific problem or issue. Participants provide their perceptions, known as “concepts.” The concepts are written down, and refined through more open-ended questions. Relationships between concepts are identified (for example causality e where one concept leads to or influences another). Concepts are considered “nodes” and the relationship between concepts are considered “links.” Links have arrowheads that show the causative direction (for example, see Fig. 2). Typically, individual maps contain up to 100 nodes and group maps that are made by

305

merging individual maps may contain as many as 800 nodes (Eden, 1988). The notion of team or group maps began in the 1990’s with the idea of helping teams negotiate consensus and commitment to a portfolio of actions (Eden, 1988). Because it makes explicit the concepts of different individuals, cognitive mapping helps facilitate decision-making by promoting a shared understanding of potential problems and design choices (Swan, 1997). With a wider understanding of the issues, negotiation can occur more easily and decision makers can jointly understand the complexity and consequences of a decision (Shaw et al., 2009). The use of a group strategy map also removes individual ownership of the issues, creating some distance to see and discuss the problem in new ways, thereby facilitating organizational change. 2.2. Review of cognitive mapping methodologies In this section we will present some of the key methodological choices in creating and analyzing cognitive maps, with an emphasis on techniques that are most likely to be of practical use in the context of human factors in manufacturing environments. This section will include: methods to elicit information; the role of the facilitator; mapping methods (software or paper and pencil); and methods for analysis and interpretation of maps. The choice of technique for any given context depends on a number of variables, including the likelihood of producing valid and reliable data, logistical considerations such as the time and extent of participation, and seniority of the participants. Other variables include the complexity of the problem, the interest of the practitioner/researcher, and the scope of analysis. It is important to consider the overall purpose and intent of the map, for example, in this case, to prioritize human factors efforts to support the organizational strategy. Note that none of the methods presume one must have a strong knowledge of the HF. 2.2.1. Choose the method of eliciting information Information can be elicited either through open-ended questions, or through pre-selected “closed” questions. Using an openended question, such as “How may one improve customer service?”, tends to result in wide-ranging and distinctive maps for any given individual. Alternatively, the researcher may use a more closed structure to provide a set of pre-selected concepts based on the literature and their domain knowledge of the situation, that the participants link or rank by importance (see Markoczy and Goldberg, 1995). One example of a closed structure is pairwise comparison, where participants make judgments of the positive or negative influence of one variable on another in a pairwise fashion (Hodgkinson et al., 2004). Another example of a closed structure is the repertory grid, an early cognitive mapping technique that involves a very structured approach for clustering and rating concepts (Eden, 1988; Swan, 1995). The advantage of closed questions is that they make merging of individual maps easier because the concepts are all similar. The disadvantages are that they do not facilitate a rich subjective reflection on the topic, and they presuppose prior knowledge of all relevant domain elements. While either open or closed questions can work, open questions allow individuals to view their responses, re-evaluate, make new links, and at times discover emergent themes that would not be otherwise captured. 2.2.2. Consider the role of the facilitator The role of the facilitator is an important consideration as it influences the mapping outcome, and it varies widely across different techniques. Some of the methods are executed exclusively by the facilitator/researcher; others are facilitator-led but

Author's personal copy

306

J. Village et al. / International Journal of Industrial Ergonomics 43 (2013) 304e313

incorporate a large degree of participant involvement; and a very few are performed with minimal involvement of the facilitator. The latter techniques involve less researcher bias e as for example, in the Self-Q Technique (Nicolini, 1999) in which participants perform self-interviews formulating their own questions with structured activities to develop their concepts and the network. These later techniques most closely resemble ethnographic methods of data collection (Nicolini, 1999). Group maps can then be built through consensus activities or the use of software that combines individual maps into group maps without any interpretation by a researcher (Nicolini, 1999). In techniques where the facilitator plays a very active role, they may create the map, add concepts and emergent themes, draw links, and cluster concepts together (see Oval Maps or Group Map Techniques by Tegarden and Sheetz, 2003; Robertson and Williams, 2006). The advantage of an active facilitator is that they can detect emerging concepts that the individual or group may not see. However, these methods are heavily influenced by the facilitator. In the middle ground are methods where the facilitator plays an active role, but also heavily involves the participants in negotiation of a consensus (see the Strategic Options Development and Analysis (SODA) method by Eden (1988) and Swan, 1995). 2.2.3. Choose the method of mapping Mapping is usually done actively with the participant, but can be done after an interview by coding transcripts of interviews or other text (Swan, 1995). Coding after an interview does not allow a participant to see the map and to question the links or expand on a concept as they are viewing it (Eden and Ackermann, 2004). Coding post-interview may also result in the facilitator making assumptions when drawing the links since they cannot check with the participant directly. The mapping method uses mapping software or paper and pencil. The extensive work by Eden et al. (1992), Eden, 1992, Eden and Ackerman (2004) and Ackermann et al. (1992) describe in detail the paper and pencil process for eliciting concepts and drawing maps. They suggest mapping in layers or hierarchies with the goals at the top, the strategic decisions in the middle and the potential options at the bottom (as is shown in Fig. 2). They suggest evaluation of opposite extremes of concepts (i.e. positive and negative), and distinguishing between actors (persons) and actions (verbs) in the codes. The advantage of paper and pencil techniques is that they require minimal technology and can easily be executed during interviews, with the participant seeing the “whole” map develop visually as they actively assist in its creation. This process helps the participants see new links and concepts they might not otherwise have noticed. It is also considered ethical in that participants can see what is recorded (Brown, 1992). Several types of software have been developed to aid both the map drawing and, especially, the analytical processes (see Lee et al., 1992 for Collective Cognitive Mapping System (COCOMAP), Ackermann, et al., 1992 for Decision Explorer, Sheetz, et al., 1994 for VisionQuest and Eden, 1988 for COPE software). In simpler programs, the facilitator uses the software to draw the map during individual or group sessions. They can later overlay individual maps to create a group map, or they can analyze maps in numerous other ways. Collectively, computer-based systems that combine hardware, software and procedures to structure group activities are known as Group Support Systems (GSS). Some GSS require group members to simultaneously work on separate computers and the systems prompt members to agree to common themes and links. Other GSS tools include electronic brainstorming, point allocation, and voting and ranking systems (Sheetz et al., 1994). The software then merges the complex maps and facilitates numerous analysis techniques. For example, rules can establish cut-offs for merging

concepts and relationships from multiple maps, such as minimum level of agreement of concepts. There is tremendous advantage in using software to analyze the information statistically. However, while such analyses may be interesting to the researcher, the data may not be as useful for the participants, or necessary for problem solving in the organization. Note that analysis methods will be discussed in Section 2.2.5. 2.2.4. Methods for creating group maps There are a range of methods for creating group maps that include: maps made by the facilitator; by the facilitator and participants together; or by participants with little input from the facilitator. Group maps may also be created without first making individual maps. The first type, where the facilitator alone merges individual maps into a group map, includes examples such as the Etiograph, Congregate Map and Group Map techniques (Swan, 1995; Tegarden and Sheetz, 2003; Robertson and Williams, 2006). In the Congregate Map technique, for example, only the nodes that are common to each individual’s map are included. This method has been criticized for researcher bias, but has the advantage of likely being quicker and involving less participants’ time. However, the final group map may not promote change or action by the group if participants are not involved in its creation. In the second type of method, where the facilitator develops the merged map together with participants during a workshop, the level of consensus must be considered. Langfield-Smith (1992) reported that reaching consensus on all concepts is time consuming, and Nicolini (1999) found it may provoke disagreement over the meaning of the nodes. However, if the maps are accepted too readily by participants, they may contain generic statements that may not be as useful. The third type of method, which relies heavily on participants deriving the map, includes, for example, Strategic Options Development and Analysis (SODA) (Swan, 1997). Software is used to retain all individual differences in individual maps. Negotiation among participants to determine the final shape of the map leads to a richer understanding of complex issues. This type of merging process, while time consuming, allows common features of each individual map to be highlighted while retaining the idiosyncratic beliefs of individuals (Swan, 1997). Finally, a group can create a shared map without first deriving individual maps (see Sheetz et al., 1994). Sheetz et al. (1994) reported that the group procedure, which took almost 4 h, was much more time efficient and produced the same overall result as conducting individual maps that were then merged. Robertson and Williams (2006) also suggests using a direct group approach if time is limited or if it is essential that all parties identify with, and buy in to, the output. The use of individual maps, however, is suggested when individual concepts are of interest to the facilitator and when participants come from different levels or departments within the organization, or where there may be disagreement between participants. 2.2.5. Methods to analyze and interpret a cognitive map A number of analysis techniques will be discussed in this section to provide an overview of options available for both individual and group maps. The case study that follows will give examples to help illustrate some of the methods. Maps are analyzed in three ways: the number of concepts and links; the shape and structure; and the content. A map’s complexity, reflected by the number of concepts, provides an indication of the person’s knowledge or expertise on a topic (Novak and Canas, 2008). While a greater number of nodes generally represents greater mastery, expert knowledge, or

Author's personal copy

J. Village et al. / International Journal of Industrial Ergonomics 43 (2013) 304e313

complexity of the situation, the method of elicitation and length of interview time can influence the number of nodes (Eden, 1992). In addition to the number of concepts, the ratio of links to concepts also reflects knowledge and expertise. Eden (1992) reports that typical ratios of the number of links to the number of concepts range from 1.15 to 1.2, with higher ratios depicting greater complexity and expert knowledge. Regarding shape and structure, cognitive psychologists suggest that the “essence of knowledge is structure” (Ruiz-Primo, 2004). Therefore, as a person’s knowledge increases through learning, training, or experience, the elements of their knowledge become increasingly connected; this can be revealed in their cognitive map. Work by Hay and Kinchin (2006) in the educational field suggest that more expert maps resemble networks with high numbers of interconnections, compared with those of novices that are linear or spoke-shaped (with numerous concepts coming from a singe concept). One can also explore the ratio of heads to tails. Concepts that are “heads” are at the top of the map with concepts leading into them. “Tails” are at the bottom of the map and represent the action items leading upward. A high ratio of heads-to-nodes can indicate multiple potentially conflicting objectives or goals, which in turn can indicate higher complexity of the problem. A high ratio of tails-to-nodes, or relative flatness of the shape, can indicate a large range of possible options to achieve a particular goal. The content of the map can be explored by looking for the most frequently linked concepts (known as domains) or most central concepts (calculated with a distance decay function for links successively further from the central concept). Central concepts reflect those with highest overall networks, or perceived “influence”. Concepts identified by domain or centrality analysis can then be remapped as smaller maps to improve comprehension of concepts and their associations. Map content can also be explored with cluster analysis to identify groupings of concepts that appear to stand-alone with respect to other groupings. Analysis software looks at each node and the immediate context of that node to determine the similarity rating (mathematically known as Jaccard co-efficient). A map with few clusters indicates less complexity as it cannot be simplified and broken apart. Finally, content can be analyzed qualitatively by following paths either from a strategic goal to a final action item, or vice verse. Software can list the various “explanations” (concepts that explain or lead into the given concept), or “consequences” (concepts that arise as a result of a given concept). In a highly networked map, there may be hundreds of these for central concepts. Particularly informative paths are the causal loops where concepts connect circularly. Both virtuous and vicious loops e indicating growth, decline, or feedback control e may be studied as opportunities for action. 3. Illustrative example of cognitive mapping 3.1. Purpose The purpose of the following illustrative example is to demonstrate the selection and use of one of the cognitive mapping techniques and evaluate its potential as a means to improve understanding of the relationship between human factors and an organization’s strategic goals. We are less interested here in the content or results of the map, than in the how the tool can be used by HF Engineers and others in industrial applications. An individual interview was conducted with one of the co-authors of this paper (WPN), who has 20 years of industrial experience. To evaluate the applicability of the tool for our larger case study in an electronics firm, the individual responded to the mapping exercise from the perspective of a manufacturing company.

307

3.2. Case study: methods Our goal was to choose a method that would be both time- and resource-efficient for the illustrative example since our broader aim was to use the method with a group of senior directors of a company. We therefore chose an interactive paper-and-pencil individual mapping technique structured with strategic goals, main goals and actions (Eden et al., 1992). The participant was provided with general instructions about how the mapping would be executed interactively during the interview. The open-ended question posed was “How can integrating human factors into production system design processes help achieve strategic goals of an organization?” Both an open-ended question and paper-and-pencil mapping were chosen so that the participant could explore their perceptions and see their concepts emerge in real time, with the potential to discover new or emergent links among the concepts. As the participant responded, concepts (short phrases or keywords in the participant’s words) were written on the paper and put in boxes. The participant was asked to verify if a concept was a strategic goal, in which case it was placed near the top of the paper or a main goal (in the middle). Action items populated the bottom of the map. The drawing and linking of concepts continued more or less uninterrupted. If unsure about the directional nature of the linkage for a given concept, the participant was prompted with “what does this lead to?” or “what causes this?” The participant was encouraged to observe the mapping process and suggest new links between concepts. The interview was scheduled for one hour, and it was digitally recorded. If a concept was missed during the interview, it was added to the map later upon reviewing the digital recording. If the facilitator was unsure of any concept or linkages, clarification was sought from the participant. For data analysis, each concept and link from the hand-drawn map was entered into Decision Explorer software (Banxia Software Ltd. 2002, U. of Strathclyde). The analyst (JV) carefully compared the computer-drawn map with the hand-drawn map and the digital recording to ensure all concepts and links were included. Strategic goals and main goals were given different box fonts and colors to make them easier to identify. Where possible, concepts were arranged to minimize linking lines that crossed; this improves legibility of the map. Initial analysis of the map included a count of concepts, heads, tails, and loops, and listing of domain and centrality scores. The most central concepts were re-mapped into smaller maps. Maps were discussed with the participant to review insights regarding both content of the map and process of its creation. 3.3. Case study: results The one-hour interview produced a cognitive map with 71 concepts. Few concepts were missed (10 links Number of clusters Number of loops

71 123 1.7 8 21 0.11 0.296 1 4 6 55 2 37

large number of loops (37) illustrate that there are many possible actions (tails) and numerous routes to achieving the goals. As shown in Fig. 1, and Table 2 the participant highlighted one strategic goal, four main goals and six sub-goals. To demonstrate the results of the analysis, Table 3 lists the ten highest central concepts (defined in the methods) with their scores, and the number of concepts linked to each. Not surprisingly, some of the most central concepts are strategic, main or sub-goals (5 of 10), since one would expect many concepts and links would lead to these. Of more interest in the centrality listing are those concepts that are central but not goals (such as “define chunks of work that can be run in parallel rather than serial”, and “improve work-rest recovery” e number 56 and 52, respectively). A re-drawing of the central concept “Improve work-rest recovery” is shown in Fig. 2. Smaller maps, such as this, are more easily read and comprehended than the full map. This is helpful for identifying action items. It helps when reviewing and interpreting the smaller maps, such as Fig. 2, to start with the central concept, in this case “Improve work-rest recovery” (box 38 in Fig. 2). Arrows leading out of this concept highlight the consequences of improving work-rest recovery, which can be sub-goals such as “improved quality” (6), or main goals such as “Increased yield and throughput”. The three arrows leading in to “improve work-rest recovery” (38) can then be explored as ways to achieve this concept, for example, “reduce short cycle jobs” (39). One can also trace the paths leading into concepts; these show various routes to achieve improved work-rest recovery. Both Figs. 1 and 2 show numerous loops, but these are difficult to disentangle when viewing a large map. The Decision Explorer software identified 37 loops, and these can be drawn individually and explored in more detail. Fig. 3 shows an example of one such reinforcing loop that contains several loops within it. Loops in Fig. 3 illustrate the highly interconnected and reinforcing nature of the concepts as understood by this participant. Taking action towards any one of these concepts, such as “reducing

Table 2 List of strategic goal, main goals and sub-goals. Strategic goal

Release new product on regular basis

Main goals

Increase product variety Improve rapid set-up Improve rapid ramp-up operationally Increase yield and through-put Increase quality Improve changeover of assembly line (ramp down) Improve model for setting up assembly Improve set-up of assembly Improve design for assembly to improve product design Improve material supply strategy

Sub goals

309

physical forces” or “reducing short cycle jobs” will have benefits in several areas. The analysis of loops is a particular advantage of the cognitive mapping technique, as can be seen even from a single person example. The knowledge contained in a loop is often not consciously evident to the participant, but is brought out through the mapping process. Loops can be powerful for beneficial change since it leads to actions that reinforce other actions. This demonstrates viable results from the mapping tool that can lead to specific improvement projects. 4. Discussion This paper provides a comprehensive review of methodological options, not available elsewhere in the literature, for those who may be interested in conducting cognitive mapping in the HF domain. It also shows the outputs from one time- and resourceefficient method of paper and pencil mapping with the assistance of software for analysis. The illustrative example demonstrates how cognitive mapping, if performed with managers or engineers, can provide a unique macroergonomic tool for diverse actors to understand HF perspectives and strategically align these with an organization’s goals (c.f. Dul and Neumann, 2009). Based on the methodological review and illustrative example, this paper has provided the method development work for a larger trial underway with a group of senior management of an electronics company. The goal of the trial is to understand management perceptions of HF, see how they align HF with their strategic goals, and help identify specific HF initiatives as part of a multi-year collaboration to increase HF application. In management research, cognitive mapping at the individual level has been used to illustrate the idiosyncratic knowledge that the individual manager possesses (Tegarden and Sheetz, 2003; Langfield-Smith, 1992). The individual mapping exercise demonstrated graphically in this paper made apparent the relationships between HF and the company’s strategic goals, as understood by the participant. This can be very valuable for a HF Engineer, or other actor, seeking to understand the perspective of managers and engineers in an organization. As well as revealing knowledge visually, in some cases it can reveal gaps in knowledge. Similar techniques have been used to explore students’ knowledge base in fields such as science, engineering, nursing, teaching, business and the arts (Hay et al., 2008). Further advantages of mapping as a technique are that the mapping process stimulated deep responses from the participant and is a useful way to understand concept linkages, to correct misunderstandings, and to make connections between concepts that may otherwise not be apparent. The visual nature of mapping allows for evidence checking and can more fully lead to saturation on a topic than other methods. It may also lead to enhancing or expanding the conceptual framework by raising other questions, and revealing emergent themes (Daley, 2004). Maps are also a basis for self-reflection that produces learning. When merged group maps are discussed with a management team, participants compare their concepts with others (share their cognitive maps), which updates their own map to facilitate learning (Fiol and Huff, 1992). It has been suggested that learning takes place at the intersection between maps and mental models (Moreroft, 1992). Using cognitive mapping, rather than conventional meetings, helps managers think more critically, broadly, and in-depth, while recognizing the contributions of others and seeing the broader environment (Shaw et al., 2009). In strategic management and operational research, mapping has been used with individuals and groups to understand or solve complex or “messy” problems such as a business decline, rationale for joint ventures, and management perceptions of competitive positioning, as well as in group model building for dynamic decision-making (Moreroft, 1992;

Author's personal copy

310

J. Village et al. / International Journal of Industrial Ergonomics 43 (2013) 304e313

5 Rapid Ramp-Up (Operationally)

7 Increased Yield and throughput

8 Improve changeover of Assembly Line (Ramp down) 9 Faster to achieve design

11 Increased flexibility

12 Improved model for setting up assembly (consider options)

21 less errors 6 Improved Quality

33 Reduced distraction 38 Improve work-rest recovery

13 Improved set-up of assembly 68 Improve design of off-time tasks

32 Reduced worker pain and fatigue

37 Use good processes 31 Reduce control and dexterity demands

34 Define chunks of work that can be run in parallell rather than seriel

58 increases job scope

57 Improves effort reward balance 44 Reduce reaches 39 Reduce short cycle jobs

36 Increase variety with workstation level changes

51 Improved psychosocial factors

53 Increased control and reduced demands

46 Reduce awkward postures 70 less monotony 40 Less attention focus rather than assembling whole product

41 Increased motivation

Fig. 2. Map of central concept: “Improve work-rest recovery” (concept #38).

Swan, 1995; Langfield-Smith, 1992). Aligning HF considerations with organizational strategies can also be considered a “messy” problem. Depersonalizing individual concepts into a group map with the goal of problem solving has been used successfully to help individuals “change their mind” in a way that is face-saving (Swan, 1997). We anticipate that cognitive mapping can also be used as a bridge or boundary object (Swan, 1995; Langfield-Smith, 1992). Broberg (2010) describes boundary objects as concrete or abstract artefacts that facilitate communication between different social groups. He reviewed various boundary objects used by HF

Table 3 List of 10 highest central concepts, scores and number of linked concepts. Highest central concepts

Score

Number of linked concepts

Improved set-up of assembly Improved model for setting up assembly Define chunks of work that can be run in parallel rather than seriala Improve quality Improve work-rest recovery Improve design for assembly to improve product design Improve psychosocial factors Increase variety with workstation level changes Improve error detection Rapid Ramp-up Operationally

29 29 28

57 58 56

28 25 25

54 52 51

23 23 23 23

43 44 50 50

a Concepts in italics were not identified as strategic, main or sub-goals by the participant.

specialists with engineers or architects to facilitate communication of knowledge across the two domains about a specific design problem (Broberg, 2010). Since human factors departments are often separate from senior management, such a boundary object may provide a common ground for discussion of the benefits of HF. The mapping tool has been recognized as a facilitator of systems level thinking, since it goes beyond simple cause and effect (Shaw et al., 2009). Such systems level thinking is consistent with the goals of human factors and macroergonomics. In a research context, the mapping tool could potentially be used as a basis for a simulation model to explore the most effective action plans for integrating human factors into the business strategies. A similar qualitative systems dynamic simulation model was developed to investigate attribution of overrun costs in a large engineering project (Howick et al., 2008). While cognitive mapping is common in operations and management research and the educational domain, its use has been limited in HF. Only a handful of HF studies report using a mapping tool. McNeese et al. (1995) reported a study that used mapping to improve elicitation of knowledge about users, tasks and their environment, and the U.S. Navy used mapping to perform job task analyses of duties, and to brainstorm challenges (Dumestre, 2004). Mapping has also been used to document and preserve institutionalized expert knowledge in the nuclear power industry and for training and learning purposes (Coffey et al., 2004; Basque et al., 2004). Mapping has also been discussed in the qualitative research field as a strategy to reduce large volumes of data and allow visual identification of themes and patterns (Daley, 2004). An average 20 page interview can be expressed in a single large map (Daley, 2004).

Author's personal copy

J. Village et al. / International Journal of Industrial Ergonomics 43 (2013) 304e313

31 Reduce control

311

51 Improved

and dexterity

psychosocial factors

demands

32 Reduced worker pain and fatigue

36 Increase variety with workstation level changes

30 Reduced physical

38 Improve work-rest

forces

recovery

46 Reduce awkward postures

39 Reduce short cycle jobs

Fig. 3. Example of a loop revealing reinforcing characteristics of human factors.

The HF literature, especially in macroergonomics and participatory ergonomics, suggest that management support and involvement are critical for success of HF in organizations (Driessen et al., 2010; Dixon et al., 2009). However, very little guidance to gain this support and involvement is available. All too often, ergonomists attempt to ”convince” or “sell” HF to management, generally appealing to injury reduction or cost-benefit, as a way to gain their commitment (Goggins et al., 2008). This is difficult and often unsuccessful. Cognitive mapping may be a tool that can facilitate managerial support since one of its strengths is that it contains only the concepts expressed by the map “owner” or manager e it is their worldview on a topic. Through mapping, the manager reveals the links they themselves perceive between HF and strategic goals, and thereby take ownership of the information. The technique also helps identify possible HF actions that support strategic goals of the organization and will have the greatest impact, again from the perspective of the manager. As such, the tool is “participatory” and helps a manager focus on human factors at the “macro” level to realize the potential links it has to organizational strategies. As mentioned in Section 2.1, when mapping is used in a group to increase understanding and negotiate a portfolio of actions, the tool promotes organizational change. We anticipate that the tool may leverage HF support and action if a managerial group shares their common perceptions about HF, and agrees to take actions to promote HF. As with any tool, cognitive mapping has limitations. Maps are shaped by the method used to build them (Swan, 1997). Some methods can take considerable time and are labor-intensive, especially if a group consensus is desired (Hodgkinson et al., 2004). Langfield-Smith (1992), in a case study with firefighters, reported that significant time was wasted due to disagreements about different meanings and boundaries of the elements, possibly because there was not a high level of shared language between members of the group. Based on their difficulties, Langfield-Smith (1992) suggested that a collective map may not be an enduring phenomenon but rather transitory “collective cognitions” subscribed to in varying degrees at a particular point in time during “collective encounters”. Some authors suggest that individual maps give richer understanding of a subject area and avoid possible takeover of forceful

personalities or “group think” that may occur when creating a map in a group setting (Robertson and Williams, 2006). On the other hand, others argue that cognitive maps of key individuals do not necessarily relate to decisions made at an organizational level, and therefore group maps with shared beliefs may be more likely to guide decision making at the organizational level (Swan, 1997). To further confound matters, O’Connor and Johnson (2004) argue that a team map that results from interaction with the team can change how individual’s think. Clearly, there are advantages and disadvantages to different methods that must be aligned with the purpose for performing cognitive mapping. Swan (1995) criticizes available methods for mapping, suggesting that most are complex and time-consuming and of little use to practitioners compared with researchers. Extensive analyses may be well beyond what is practical in an organizational setting. The practitioner focused on time and effectiveness may choose more direct individual or group approaches and simpler analyses in order to move toward action to solve the problem or further the HF agenda. When considering the question of how to judge the usefulness of the map, Fiol and Huff (1992) suggest the map should provide “sufficient understanding for effective action”. For most HF Engineers, the goal of mapping will be to agree to action steps in accordance with an objective, such as reported by Robertson and Williams (2006) when looking for recommendations about how to proceed with huge delays in a software implementation project. While it can be argued that a cognitive map of an individual or group may only be valid for a particular point in time, it may be that the actions identified through the mapping exercise stay relevant, despite changes in perceptions or management. However, for mapping to result in tangible action and therefore reach its potential as a tool for organizational change or macroergonomics, the group must be prepared to act on the action items identified by the cognitive mapping exercise. In other words, the purpose of cognitive mapping and the process of change should be made clear to participants ahead of time and agreement should be reached as to the steps that follow from mapping. If the purpose is to explore knowledge, to compare knowledge between participants, or to compare knowledge before and after a change, then mapping may provide insight but not necessarily lead to specific action. Alternatively, if the purpose is to solve a specific problem, or negotiate a

Author's personal copy

312

J. Village et al. / International Journal of Industrial Ergonomics 43 (2013) 304e313

solution, or improve HF activities in an organization, then mapping can reveal actionable steps, but there still needs to be a commitment to making the changes. 5. Recommendations Based on the methodological review and our illustrative example, we suggest the following recommendations for the practical use of individual cognitive mapping by HF Engineers with managers and others in organizations to link HF considerations to strategic goals:  conduct individual paper-and-pencil mapping interviews of one hour duration with managers and others to optimize information in a reasonable time period;  use an open-ended question to gather perceptions about HF, and how it links to strategic goals in the organization;  consider recording the interview so information can be checked without needing to re-check with the manager;  use software to re-draw the maps and to disentangle the concepts;  run simple analyses to reveal the main strategic goals and central concepts from the individual maps;  Analyze the most central concepts for discussion (highest number of links);  Investigate the types of actions (tails) that include HF and link to strategic goals;  Analyze loops with reinforcing concepts as they have the potential for ongoing positive feedback; and  Compare human factors perceptions between different individuals. Based on the illustrative example, we propose that individual mapping with managers in a manufacturing organization will provide untapped perceptions about HF and their link with strategic goals. Based on our literature review, we further suggest merging maps into a group map based on a common goal. 5.1. Next steps In our larger trial underway with an electronics company, we will perform mapping for each individual in a management team, which may reveal both individual beliefs with respect to HF and differences in knowledge sets and perspectives. We then propose combining single maps of managers into a group map, as discussed methodologically in Section 2.2.4. Our main goal is to generate shared understanding and action to promote the uptake of HF in their design of assembly systems. We will then evaluate the utility of the merged map in a workshop with managers for:  sharing perceptions about HF with the group to facilitate discussion and learning;  determining the most central connections between human factors and strategic goals for this organization; and  agreeing to a portfolio of potential HF actions based on central concepts and paths in the group map and discussions in the workshop. The outcome of the workshop is anticipated to be threefold: human factors specialists gain a critical understanding of the perceptions managers have of strategic priorities and HF; managers increase their understanding of how human factors can help them achieve their strategic goals; and actionable initiatives can result from the workshop that further strengthen the connection between HF and the strategic goals.

6. Conclusions This paper presents a tool called cognitive mapping, from operations and management research, that has the potential to help Senior Managers and others understand how human factors contributes to strategic goals in their organizations. The various methodological and analytical options for cognitive mapping are reviewed in this paper to help the HF Engineer (or others) choose mapping methods and analyses appropriate for their specific context. We recommend further application and testing of this tool. The maps have the potential to facilitate participatory human factors knowledge sharing at a systems or macroergonomic level, and can lead to actionable steps to further integrate human factors into organizations. This tool has the potential to change the human factors emphasis in organizations from one of short-term costbenefit savings to enhancing long-term strategic objectives. This may open up new opportunities for the application of HF in industrial systems. Acknowledgements The authors are grateful for the funding supplied by the Workplace Safety Insurance Board of Ontario. References Ackermann, F., Eden, C., Cropper, S., 1992 (13-15th April). Getting Started with Cognitive Mapping. University of Warwick, pp. 65e82. Basque, J., Imbeault, C., Pudelko, B., Leonard, M., 2004. Collaborative knowledge modeling between experts and novices: a strategy to support transfer of expertise in an organization. In: Canas, A.J., Carff, R., Suri, N., Lott, J., Gomez, G., Eskridge, T.C., Arroyo, M., Carvajal, R. (Eds.), Proc. of the First International Conference on Concept Mapping. (Pamplona, Spain). Broberg, O., 2007. Integrating ergonomics into engineering: empirical evidence and implications for the ergonomists. Human Factors and Ergonomics in Manufacturing 17 (4), 353e366. Broberg, O., 2010. Workspace design: a case study applying participatory design principles of healthy workplaces in an industrial setting. International Journal of Technology Management 51 (1), 39e57. Brown, S.M., 1992. Cognitive mapping and reprtory grids for qualitative survey research: some comparative observations. Journal of Management Studies 29 (3), 287e307. Coffey, J.W., Eskridge, T.C., Sanchez, D.P., 2004. A case study in knowledge elicitation for institutional memory preservation using concept maps. In: Canas, A.J., Carff, R., Suri, N., Lott, J., Gomez, G., Eskridge, T.C., Arroyo, M., Carvajal, R. (Eds.), Proc. of the First International Conference on Concept Mapping. (Pamplona, Spain). Daley, B.J., 2004. Using concept maps in qualitative research. In: Canas, A.J., Carff, R., Suri, N., Lott, J., Gomez, G., Eskridge, T.C., Arroyo, M., Carvajal, R. (Eds.), Proc. of the First International Conference on Concept Mapping. (Pamplona, Spain). Dixon, S.M., Therberge, N., Cole, D.C., 2009. Sustaining management commitment to workplace health programs: the case of participatory ergonomics. Resations Industrielles/Industrial Relations 64 (1), 50e64. Driessen, M.T., Groenewoud, K., Proper, K.I., Anema, J.R., Bongers, P.M., van der Beek, A.J., 2010. What are possible barriers and facilitators to implementation of a participatory ergonomics programme? Implementation Science 5 (64), 1e9. Drury, C.G., 2000. Human factors and quality: integration and new directions. Human Factors and Ergonomics in Manufacturing 10 (1), 45e59. Dul, J., Neumann, W.P., 2009. Ergonomics contributions to company strategies. Applied Ergonomics 40, 745e752. Dumestre, J., 2004. CMap tools software to assist in performing job task analysis. In: Canas, A.J., Carff, R., Suri, N., Lott, J., Gomez, G., Eskridge, T.C., Arroyo, M., Carvajal, R. (Eds.), Proc. of the First International Conference on Concept Mapping. (Pamplona, Spain). Eden, C., 1988. Cognitive mapping. European Journal of Operational Research 36, 1e13. Eden, C., 1992. On the nature of cognitive maps. Journal of Management Studies 29 (3), 261e265. Eden, C., Ackermann, F., 2004. Cognitive mapping expert views for policy analysis in the public sector. European Journal of Operational Research 152, 615e630. Eden, C., Ackermann, F., Cropper, S., 1992. The analysis of cause maps. Journal of Management Studies 29 (3), 309e324. Fiol, C.M., Huff, A.S., 1992. Maps for Managers: where are we? Where do we go from here? Journal of Management Studies 29 (3), 267e286. Genaidy, A.M., Sequeira, R., Rinder, M.M., A-Rehim, A.D., 2009. Determinants of business sustainability: an ergonomics perspective. Ergonomics 52 (3), 273e301.

Author's personal copy

J. Village et al. / International Journal of Industrial Ergonomics 43 (2013) 304e313 Goggins, R.W., Spielholz, P., Nothstein, G.L., 2008. Estimating the effectiveness of ergonomics interventions through case studies: Implications for predictive cost-benefit analysis. Journal of Safety Research 39, 339e344. Hay, D.B., Kinchin, I.M., 2006. Using concept maps to reveal conceptual typologies. Education þ Training 48(2/3), 127e142. Hay, D., Kinchin, I., Lygo-Baker, S., 2008. Making learning visible: the role of concept mapping in higher education. Studies in Higher Education 33 (2), 295e311. Hendrick, H.W., 2008. Applying ergonomics to systems: some documented “lessons learned”. Applied Ergonomics 39, 418e426. Hodgkinson, G.P., Maule, A.J., Bown, N.J., 2004. Causal cognitive mapping in the organizational strategy field: a comparison of alternative elicitation procedures. Organizational Research Methods 7 (1), 3e26. Howick, S., Eden, C., Ackermann, F., Williams, T., 2008. Building confidence in models for multiple audiences: the modelling cascade. European Journal of Operational Research 186, 1068e1083. Jensen, P.L., 2002. Human factors and ergonomics in the planning of production. International Journal of Industrial Ergonomics 29, 121e131. Kelly, G.A., 1955. The Psychology of Personal Constructs: a Theory of Personality. Norton, New York. Kirwan, B., 2000. Soft systems, hard lessons. Applied Ergonomics 31, 663e678. Langfield-Smith, K., 1992. Exploring the need for a shared cognitive map. Journal of Management Studies 29 (3), 1e20. Lee, S., Courtney Jr., J.F., O’Keefe, R.M., 1992. A system for organizational learning using cognitive maps. Journal of Management Science 20 (1), 23e36. Markoczy, L., Goldberg, J., 1995. A method for eliciting and comparing causal maps. Journal of Management 21 (2), 305e333. McNeese, M.D., Zaff, B.S., Citera, M., Brown, C.E., Whitaker, R., 1995. ADADAM: eliciting user knowledge to support participatory ergonomics. International Journal of Industrial Ergonomics 15, 345e363. Miles, B.L., Swift, K., 1998. Design for manufacture and assembly. Manufacturing Engineer 77 (5), 221e224. Moreroft, J.D.W., 1992. Executive knowledge, models and learning. European Journal of Operational Research 59, 9e27. Neumann, W.P., Edman, M., Winkel, J., 1999. Integrating ergonomics into production system development e the volvo powertrain case. Applied Ergonomics 40, 527e537. Nicolini, D., 1999. Comparing methods for understanding organizational cognition. Organization Studies 20 (5), 833e860.

313

Novak, J.S., Canas, A.J., 2008. The theory underlying concept maps and how to construct and use them. In: Technical Report, Institute for Human and Machine Cognition Cmap Tools. Florida Institute for Human and Machine Cognition. www.ihmc.us. O’Connor, D.L., Johnson, T.E., Khalil, M.K., 2004. Measuring team cognition: concept mapping elicitation as a means of constructing team shared mental models in an applied setting. In: Canas, A.J., Carff, R., Suri, N., Lott, J., Gomez, G., Eskridge, T.C., Arroyo, M., Carvajal, R. (Eds.), Proc. of the First International Conference on Concept Mapping, pp. 1e7. Pamplona, Spain. Perrow, C., 1983. The organizational context of human factors engineering. Administrative Science Quarterly 28, 521e541. Robertson, S., Williams, T., 2006. Understanding project failure: using cognitive mapping in an insurance project. Project Management Institute 37 (4), 55e71. Ruiz-Primo, M.A., 2004. Examining concept maps as an assessment tool. In: Canas, A.J., Carff, R., Suri, N., Lott, J., Gomez, G., Eskridge, T.C., Arroyo, M., Carvajal, R. (Eds.). (Pamplona, Spain). Seim, R., Broberg, O., 2010. Participatory workspace design: a new approach for ergonomists? International Journal of Industrial Ergonomics 40 (1), 25e33. Shaw, D., Eden, C., Ackermann, F., 2009. Mapping causal knowledge: How managers consider their environment during meetings. International Journal of Management and Decision Making 10 (5/6), 321e340. Sheetz, S.D., Tegarden, D.P., Kozar, K.A., Zigurs, I., 1994. A group support systems approach to cognitive mapping. Journal of Management Information Systems 11 (1), 31e58. Swan, J.A., 1995. Exploring knowledge and cognitions in decisions about technological innovation: mapping managerial cognitions. Human Relations 45 (11), 1241e1271. Swan, J., 1997. Using cognitive mapping in management research: decisions about technical innovation. British Journal of Management 8, 183e198. Tegarden, D.P., Sheetz, S.D., 2003. Group cognitive mapping: a methodology and system for capturing and evaluating managerial and organizational cognition. Omega 31, 113e125. Waterson, P., Kolose, S.L., 2010. Exploring the social and organizational aspects of human factors integration: a framework and a case study. Safety Science 48, 482e490. Zink, K.J., Steimle, U., Schroder, D., 2008. Comprehensive change management concepts. Development of a participatory approach. Applied Ergonomics 38, 527e538.