Factors Affecting Technological Trouble Shooting Skills
Randall T. MacPherson
Boeing Rotorcraft Company
How problem-solving abilities are developed, and moreover, what characteristics can be used to predict problem-solving skills has intrigued theoretical and practical researchers for years. While there is some evidence that certain attributes (e.g., cognitive knowledge and critical thinking skills) may be linked with certain problem-solving skills, these conclusions have been largely derived from controlled, clinically-based studies (D'Zurilla & Maydeu-Olivares, 1995; D'Zurilla, & Nezu, 1990; Holyoak, 1995). Few studies have explored the predictive relationship of known factors involved in problem- solving and problem-solving skills in (simulated) authentic or real-life workplace settings. Much remains to be learned about the characteristics of one unique subset of general problem-solving (i.e., near transfer troubleshooting technological problem-solving). Near transfer, in this context, is defined as the ability to apply previously learned knowledge and skills to new situations and contexts that are similar to those in which the learning originally occurred (Johnson & Thomas, 1994). Further, there is a need to explore why some technicians appear to be able to draw upon an internal, heuristic set of problem- solving tools with subsequent successful transfer, (i.e., the expert problem-solver) while others cannot.
A review of the literature suggests that both the employee and the employer benefit from identifying and assessing the extent and degree of sophistication of employee problem-solving skills (Drucker, 1993; Nichols, 1994). For example, while it may be unrealistic for employees to keep current with all technological advancements, it is feasible to apply generalizable (contextually) near transfer technological problem-solving skills to aid in bridging the gap between changing system specifications and effective fault diagnosis (Johnson & Thomas, 1994).
The identification and subsequent reinforcement of characteristics associated with successful technological problem-solving are particularly important to the industrial teacher educator. Studies have concluded technological problem-solving is a key tenet of higher order thinking (Lavoie, 1991; Waetjen, 1989) and that technological problem- solving is, by definition, rooted in real-life or authentic domains (Custer, 1995). Technological problem-solving encourages creativity, ingenuity, and inventive thought processes. Inventiveness, creativity and the ability to "think on one's feet" are hallmark employee traits desired by employers and entrepreneurs.
The literature is filled with technical definitions largely based in the fields of psychology and mathematics. A problem exists in any situation in which there is a difficulty or uncertainty that needs creative or logical solutions. "A problem arises when we have a goal-a state of affairs that we want to achieve-and it is not immediately apparent how the goal can be obtained" (Holyoak, 1995, p. 118). Gobet and Simon (1996) describe problem-solving as simply a matter of finding the best solution to resolve a problem.
Troubleshooting Technological Problem-solving
Troubleshooting, as a subset of technological problem-solving, comprises both shared common and unique characteristics (Custer, 1995; Jereb, 1996; Johnson, 1989). All problems contain an initial or "what is" state, a solution(s) path, and a goal or "what is desired" state (Newell & Simon, 1972). Jereb (1996) indicated that, "the question of how to come from a given starting situation to a desired end situation is usually the essence of each technical problem" (p. 2). While this study was specifically designed to address constructs of technologically based problems, Jereb's statement could also easily apply to all manner of problem-solving (i.e., technological, social/personal, or natural/ecological) (Custer, 1995). Stated differently, since problem-solving is a process that encompasses all facets of human existence, personality characteristics such as perseverance, ingenuity, self confidence, and patience are common to both technological troubleshooting and general problem-solving.
Conversely, not all problems have a technological component. The drug or spouse abuser may be faced with a multitude of personal problems, perhaps partially attributable to poor selection of resolution options. A company owner may be faced with problems of meeting deadlines or payroll perhaps as a result of economic factors. In each of these instances, while a technological component could be present, the problems are primarily non-technological in nature.
Purpose of the Study
The purpose of this exploratory study was to investigate the predictive relationship among factors affecting near transfer troubleshooting technological problem-solving skills. The specific focus of the study was on a subset of technological problem-solving (i.e., near transfer troubleshooting).
The study was designed to address the following research questions:
- Do the scores from the six cognitive technical skill assessment areas (i.e., electrical, electronic, hydraulic, mechanical, pneumatic and programmable logic controllers [PLCs]) predict near transfer technological troubleshooting problem-solving skills?
- Do the number of years of technical troubleshooting experience in the "trade" predict near transfer technological troubleshooting problem-solving skills?
- Does field dependence/independence as measured by the Group Embedded Figures Test predict near transfer technological troubleshooting problem-solving skills?
- Do critical thinking skills as measured by the Watson-Glaser Critical Thinking Assessment predict near transfer technological troubleshooting problem-solving skills?
- Does confidence, approach/avoidance, and personal control as measured by the Technological Problem-solving Style Inventory predict near transfer technological troubleshooting problem-solving skills?
An exploratory correlational design was used to ascertain the relationship(s) between five predictor variables (i.e., cognitive content knowledge and style, years of technical on- the-job troubleshooting experience, critical thinking skills, and technological problem- solving style) and the criterion variable (i.e., near transfer troubleshooting technological problem-solving skills). To quantify near transfer troubleshooting technological problem- solving skills, the design employed a timed (simulated) authentic exercise where a series component and system faults were inserted into an automated manufacturing cell. In addition, a qualitative component was employed to further probe the thought processes used during critical decision-making and information processing junctures in the technological troubleshooting exercise.
Population and Sampling
The population for this study consisted of maintenance technicians employed in light manufacturing and service industries located in the mid-region of the state of Missouri. The sample population consisted of 15 maintenance technicians from representative electronic, automotive sealant, food processing, and technical service departments of major companies in the region. It is important to note that although the sample was drawn from companies that produced a variety of different products or services, each of the participants shared many common job responsibilities. For example, each was responsible for "real time" production line troubleshooting tasks such as: (a) electrical, electronic, mechanical, fluid power repair and maintenance; (b) PLC programming; (c) removal and repair of production line components; (d) system diagnosis for optimum speed and feed settings; and (e) calibration of timers, relays, and various controlling sensors. Prior to the administration of the predictor variable testing instruments, the participants received a written invitation outlining the research design along with the anticipated time frame for data collection. The letter emphasized that an individual's performance results would be kept strictly confidential and would not be linked to compensation or advancement within their company. In addition, the written invitations stated the date, time, and location for the participants to meet to ask questions and to begin the data collection. Confidentiality of the results was ensured by the direct delivery of individually sealed envelopes to each participant by the researcher.
Data collection began with the administration of the five standardized "pencil and paper" type, predictor variable assessments. In order to maximize continuity of thought with the limited time segments available with each technician, given the assessments were divided and administered in the following order (a) PSI-TECH, (b) years of experience, (c) GEFT, (d) WGCTA, and (e) the six cognitive technical knowledge assessments in random order. The assessments were administered either on-site at each company location; or the university campus. This phase of data collection required several sessions and typically required 7-8 hours per participant.
In order to assess near transfer troubleshooting technological problem-solving ability, a critical-incident, technique-based instrument was developed, entitled the Near Transfer Technological Problem-solving Skills Assessment (NTTPSSA). Similar to the format developed by Dyrenfurth, Custer, Loepp, Barnes, Iley, and Boyt (1993), this model was chosen from among various other options because of "its enhanced reliability and its promise of yielding the richest descriptive, scalar information, as well as its particular suitability to...numerous parallel observable incidents/methods to identify the extent of implementation along the...continuum" (p. 63).
The NTTPSSA was designed to appraise and quantify both the cognitive and psychomotor near transfer technological problem-solving performance of maintenance technicians in a simulated workplace setting. More specifically, the instrument was designed to assess the following problem-solving dimensions: (a) sensing that a problem exists, (b) identifying and defining the problem, (c) hypothesizing and clarifying the goal of the problem situation, (d) judging if more information is needed, (e) solving single and multiple-solution problems, and (f) verifying the solutions. Based on the parameters established by Johnson (1989, p. 23) for technological troubleshooting exercises, six criteria were included in the criterion variable assessment instrument. These criteria included (a) the problem must represent an actual live system failure, (b) it must be relatively easy to insert the failure, (c) the problem must not compromise personnel or equipment safety considerations, (d) the problem should not be visually detectable, (e) no intermittent faults are permissible, and (f) single failures are preferred over multiple interconnected failures.
Building on the technological troubleshooting foundational work of Johnson (1989), the NTTPSSA rubric was divided into five key dimensions of technological problem-solving. They were (a) information acquisition, (b) information interpretation, (c) hypothesis generation, (d) hypothesis testing, and (e) hypothesis acceptance. Each problem-solver was viewed as "occupying a position on a continuum" (Johnson, 1998, p. 64) for each of the five key dimensions. Each of the dimensions was further defined by the development of key behavioral characteristics that were indicative of the positions across the five levels of near transfer technological problem-solving skills. Multiple thematic strands, unique to each key dimension, were employed to further classify the observed problem-solving characteristics within each of the five dimensions of problem-solving expertise. The common theme of each strand was to address the questions of "how well" and to "what extent" the participant displayed actions consistent within the expert to beginner problem-solver continuum range. The intent was to further assist the rater(s) in classifying the maintenance technician's actions along the near transfer technological problem-solving continuum (see Figure 1).
Validity and Reliability
To optimize the content validity, detailed reviews were solicited during the development of the NTTPSSA from several acknowledged technological troubleshooting experts and researchers including local area maintenance supervisors and experienced technicians. All of the suggested modifications were reviewed for potential inclusion prior to the pilot testing.
Reliability concerns related to the consistency of interpretation of verbal and visual protocols were addressed through the use of a second rater. During the pilot testing and throughout all of the problem-solving exercises, both the researcher and secondary rater were present. Each independently observed and recorded an assessment of the problem- solving actions via the NTTPSSA scoring instrument.
Two volunteers with "real-world" troubleshooting experience participated in the pilot- test. These volunteers were selected to represent the extremes of the problem-solving continuum (i.e., one expert and one novice). The specific goals of the pilot-test were to: (a) assess inter-rater reliability; (b) construct an instrument-specific observation, interview, and rater experience base; (c) validate and calibrate the instrument; and (d) to determine whether the NTTPSSA could reliably serve to categorize near transfer technological problem-solving actions as classified by the five problem-solving dimensions.
Data Collection Procedures
The equipment used to demonstrate near transfer troubleshooting technological problem-solving skills was the Technovate Computer Integrated Manufacturing (CIM) Cell located in the Manufacturing Systems Engineering Laboratory (MSEL) at the University of Missouri-Columbia. The cell is designed to replicate real-world systems typically found in a variety of manufacturing environments and consists of a material handling, inspection, and machining center.
In constructing the NTTPSSA, four classes of problem-solving skills were considered for potential inclusion. Each aligned well and were consistent with the six previously described troubleshooting exercise criteria. They were (a) frequently occurring and easy to diagnose, (b) frequently occurring and hard to diagnose, (c) infrequently occurring and easy to diagnose, and (d) infrequently occurring and hard to diagnose (Johnson, 1989, p. 23). Johnson (1989) suggests that the "easy to diagnosis" faults would not, by definition, differentiate between novice and skilled problem solvers and were, consequently, eliminated. Similarly, because of pattern recognition concerns, the frequently occurring faults categories were also eliminated. Problems selected for this study were from the "Difficult and Infrequent" category (see Figure 2).
In order to develop the system and component faults to be inserted in the CIM cell, a task analysis process was utilized. Three types of symptoms were chosen for the problem- solving exercise, each containing unique system and component inserted faults or failures.
They were as follows:
Robotic arm has stopped in mid-cycle (see Figure 3), Conveyor line will not start (see Figure 4), and Pallet is not passing through station #3 (see Figure 5). Prior to actual data collection, the participants were introduced to the CIM cell by way of a 10- to 15-minute overview and were permitted to view the CIM cell in a normal operational mode. A script was devised and followed to insure that all participants received identical information describing the behavior of the system, as well as the function of various components and sub-systems of the cell. Questions were encouraged and answered only during the CIM cell introductions. Each participant was subsequently taken to an adjoining room and given a one-page narrative overview outlining and describing the nature of the malfunctions. Concurrently, the faults were inserted into the CIM cell. To minimize the potential of shared information outside of the controlled problem setting, the technicians were asked to keep their experience strictly confidential. The probability of information exchange was also minimized by scheduling all members of a particular company back-to-back on the same day.
The problem scenario narrative was designed to replicate service repair requests routinely received by maintenance departments. To further replicate an authentic setting, the narrative placed the emphasis on the strategies and methodologies used to identify the inserted faults, and not on the resolution or speed in which they were found. However, due to limited availability of both the Technovate CIM cell and maintenance personnel, a maximum of 15 minutes was allotted for the identification of each inserted fault (a maximum of 45 minutes for the entire exercise). All necessary tools, test equipment, reference data (i.e., schematics, wiring diagrams, operations manuals, etc.) were supplied for use.
The technicians were encouraged to "think aloud" and verbalize their findings and strategies as they proceeded through the problem-solving exercise. These verbal protocols and actions were captured through rater observation and were augmented by the use of a stationary video camera. The video camera was mounted on a tripod on an elevated "cat walk" surrounding the CIM cell.
Immediately following each session, the raters reviewed the video tape with each participant for additional clarification and documentation of the rationale used to detect the faults. All of the sources of information (i.e., direct observations, video tape review, replies to the qualitative questions, and exit interviews) were used to complete the NTTPSSA rubric. Total data collection took approximately 40 calendar days with an average of 11 hours per participant.
Following the collection and analysis of all data, a series of group discussions were coordinated by the researcher designed to further explore the near transfer technological problem-solving self-efficacy qualitative component of the study. More specifically, the intent of these debriefing sessions (each composed of four to five technicians) was to compare the participants' rank ordering strength of the technological troubleshooting predictor variables with that of what the actual data collection revealed. To further add to the post-data collection design analysis, the participants were encouraged to voice their opinions with regard to the validity and authenticity of the general design. The group sessions were conducted at each company's site and typically lasted from 2 to 2 1/2 hours. The results of these sessions are presented in the Summary of the Findings section. The results of the five standardized "pencil and paper" type, predictor variable assessments are summarized in Table 1.
Table 1 Descriptive Statistics for Predictor Variables
Variables N Dev. Pts.
Std. Min. Max.
1. Cognitive Technical Scores Electrical 13 123 56.85 16.33 36 86 Electronic 14 86 41.50 10.06 29 61 Hydraulic 13 156 101.31 10.90 82 119 Mechanical 14 79 44.00 5.78 32 52 Pneumatic 13 131 75.62 12.89 56 99 PLCs 14 62 37.00 6.94 28 49 2.Group Embedded Figures Test Thinkning Skills Scores (Total) 14 18 12.79 4.04 6 18 3.Watson-Glaser Critical 14 80 51.64 11.57 26 69 Inference 14 16 9.00 2.35 4 13 Recognition of
14 16 11.00 4.10 3 16 Deduction 14 16 9.86 2.66 5 14 Interpretation 16 12.21 4.30 0 16 Evaluation of
14 16 9.57 4.54 0 15 4. Years of Experience 14 9.89 6.39 2 21 5 PSI-TECH Scores (Total) Problem-
14 2 63.71* 14.48 41 94 Personal Control 14 30 11.43* 4.27 5 17
*Note: Higher PSI-TECH scores indicate lower levels of problem-solving style (i.e., confidence, approach/avoidance behavior, and personal control).
The results of scores of the second phase of data collection (i.e., technological problem- solving ability as recorded and quantified by the Near Transfer Technological Problem- solving rubric) are summarized in Table 2. A maximum of 78 points was possible within the five dimensions with the scores ranging from 30 to 78.
Table 2 Near Transfer Trouble Shooting Technological Problem-Solving Range of Scores
Test Scores Frequency
70-up 5 60-69 2 50-59 3 40-49 3 30-39 1
Criterion Variable Scoring Rubric
The Near Transfer Technological Problem-solving Skills Assessment rubric proved to be particularly robust in quantifying near transfer technological problem-solving skills. Following some minor adjustments as a result of the pilot and calibration test, both raters found that the instrument provided very clear and yet flexible scoring options. In addition, the instrument possessed universal application scoring characteristics that would be relevant to a wide spectrum of problem-solving scenarios. For example, using the multiple strands of the information acquisition and interpretation dimensions, the raters were able to succinctly capture the consistency in fault identification as well as the strategies used in the process.
Data analysis began with a preliminary exploration of the relationships among the variables. This preliminary analysis was conducted in order to begin the process of coalescing a model that would be analyzed subsequently through regression procedures. A number of Pearson r combinations were computed to ascertain the extent of the positive or negative correlation between each (a) predictor and criterion variable, (b) predictor to predictor variable, (c) predictor variable and the five subsets (dimensions) of the criterion variable, and (d) predictor variable subset and the five subsets (dimensions) of the criterion variable.
Table 3 presents the Pearson r correlations between each of the five predictor variables and the criterion assessment scores (averaged scores of the two raters). A rank ordering of the strongest positive relationship to the least is presented. The strongest positive correlation (Pearson r = .3564) was between technological problem-solving and "years of experience." This finding is corroborated by previous studies that have concluded that a significant element in successful problem-solving is familiarity with the problem setting (Gobet & Simon, 1996; Holyoak, 1995; Johnson, 1989). It also makes intuitive sense that years of experience or familiarity with components within the system would contribute to problem solvers' ability to arrive at a successful solution. Based on the Pearson r value strength with the criterion variable and the degree of shared variance within the predictor variables, three of the five predictor variables were retained and two (i.e., field dependence/independence and technological problem-solving style) were discarded for the next level of analysis (i.e., multiple linear regression). Table 4 presents the change in the coefficient of determination when the three selected predictor variables were included in the model.
Table 3 Correlational Analyssis for Predictor and Criterion Variables
1. Years of Experience r=.3564 2. Cognitive Technical Skill Area Scores (Composite Average) r=.2120 3. W-G Critical Thinking Skills Scores r=.1692 4. Group Embedded Figures Scores r=.1466 5. PSI-TECH Scores r=.3856*
*Note: Higher PSI-TECH scores indicate lower levels of problem-solving self-confidence,
approach/avoidance, and personal control. Thus, a positive correlation indicates a negative
Based on the Pearson r value strength with the criterion variable and the degree of shared variance within the predictor variables, three of the five predictor variables were retained and two (i.e., field dependence/independence and technological problem-solving style) were discarded for the next level of analysis (i.e., multiple linear regression). Table 4 presents the change in the coefficient of determination when the three selected predictor variables were included in the model.
Table 4 Multiple Linear Regression
Criterion Variable = NTTPSSA Score Predictor Variables SEB ß B Multiple
ÆR % Shared
Years of Experience .6554 .3564 .8658 .3564 .3564 12.7% Cognitive Technical Content Knowledge (Composite Av.) .4218 .1017 .1437 .3689 .0125 13.6% Critical Thinking Skills .4577 .8209 1.1020 .6732 .3043 45.3 %
Note: SEB = Standard error of the beta. $#223; = Beta coefficients or multiple regression weights for standardized scores. B = Beta coefficients or multiple regression weights for unstandarized scores.
The cognitive technical knowledge assessments predictor variable (i.e., composite average of six assessments) also proved to be an effective predictor of near transfer technological problem-solving skills. This variable ranked second in correlational predictive strength and was only surpassed by the "years of experience" variable in forecasting near transfer technological problem-solving skills. When the dimensions of the NTTPSSA were analyzed, the strongest correlation was between cognitive technical knowledge and the information acquisition dimension. Conversely, the cognitive technical knowledge failed to correlate with hypothesis acceptance skills (i.e., hypothesis accuracy and acceptance of a correct hypothesis), which could be attributed to the abstract nature of this particular criterion subset. It is important to note that the technical knowledge predictor variable in combination with years of experience and critical thinking skills was particularly effective in predicating near transfer technological problem-solving skills (R = .6732).
Years of Experience
The years of experience variable was the strongest indicator of near transfer technological problem-solving skills. Based on the criterion variable subset correlations, this variable was the strongest for predicting success in the hypothesis acceptance area (i.e., hypothesis accuracy and acceptance of a correct hypothesis). This finding indicates that a solid base of content knowledge is necessary in order to be able to solve near transfer, troubleshooting problems.
The cognitive style (field dependence/independence) variable was not an effective predictor of near transfer technological problem-solving skills. This variable ranked fourth out of five in correlation magnitude with the criterion variable. Also, when this predictor was added to the multiple linear regression equation, the difference (i.e., _R) was negligible.
The critical thinking skills predictor variable was found to be an effective predictor of near transfer technological problem-solving skills. Ranking third in Pearson r magnitude with the criterion variable, this variable was surpassed only by the years of experience and cognitive technical knowledge predictor variables in predictive strength. Additional predictor-to-predictor Pearson r analysis detected a low shared variance between the top two ranked predictor variables. However, when combined with these two predictor variables, the combined shared criterion variance more than tripled. This finding suggests that in combination, cognitive knowledge, experience, and critical thinking ability serve as particularly strong indicators of near transfer technological problem- solving capabilities.
The technological problem-solving styles variable, as measured by the PSI-TECH, was the least important indicator of near transfer technological problem-solving skills. As was mentioned in the preceding paragraphs, higher scores on the PSI-TECH assessment denoted lower degrees of self-confidence, approach/avoidance, and personal control. Based on the relatively low correlation with the criterion variable and minimal cumulative effect of the shared predictor/criterion variance, the problem-solving style variable was excluded from the multiple linear regression analysis.
NTTPSSA Psychomotor Exercise Rater Observations
At the conclusion of each troubleshooting exercise, the raters shared their observations and discussed the potential implications of their findings. As a result of these discussions, several conclusions began to emerge. Based on some preliminary observations, a pattern of exhibited characteristics began to appear that distinguished expert from novice problem-solvers. For example, the more experienced technicians would initially "observe" the entire CIM cell prior to taking any action. This initial data acquisition phase also typically included a review of the supplied manuals, wiring diagrams, and schematics. Once these technicians began to take action, the steps taken appeared to be logical and deliberate.
During the exit interviews and during the video tape review, the expert technicians were readily able to explain why specific actions were taken. Another distinguishing characteristic of these individuals was that their initial hypothesis formulation was based on preliminary information acquisition and subsequent interpretation. In the majority of instances, the hypotheses formulated by the experts were correct. In cases where experts were incorrect, they quickly abandoned the false premise and formulated a replacement. The raters also observed that the novice technological problem-solvers exhibited many of the classic characteristics described in the literature. For example, rather than initially collecting information on the Technovate CIM cell as a whole system or consulting available resource reference material, these technicians often "jumped" into the exercise. The novice problem-solvers also tended to engage in repetitive, non-effective actions. When asked, for example, why an electrical switch or valve was repeatedly toggled throughout the troubleshooting exercise, novices were often unable to explain their actions.
Follow-up Group Discussions
A series of small group meetings (one per company) was conducted approximately two weeks following the completion of the simulation. The purpose of this component of the study was to capture some additional, reflective thinking on the problem-solving process from the practitioners' perspective. During each of the meetings, the goals of the study were shared with the participants, including a brief description of each of the study's independent variables. Prior to sharing the results of the data analysis, the group was invited (a) to speculate on "how they thought the study had turned out" and (b) to provide some rationale for their choices. The views that were presented and discussed during these sessions provided some valuable perspective from which to better understand the results of the study.
As could be anticipated, the perspectives of the various groups were diverse. In some cases, their views aligned closely with the quantitative results while, in other cases, they did not. Several themes emerged that became a valuable part of the reflective, analytical process.
Validity. One theme that emerged from the discussions had to do with the validity of the instruments used in the study. One group had some question about the use of the paper-and-pencil, multiple-choice format for assessing technical knowledge while members in another group wondered whether the Watson-Glaser was designed to measure the kind of critical thinking needed by technological trouble-shooters. It was useful to note that across the groups, there was (a) strong support for the validity of the variables selected for the study and (b) a consistent perception that the NTTPSSA represented a valid assessment of near transfer, technological problem-solving ability. Technical knowledge. To varying degrees, each of the groups recognized the critical importance of the technical content knowledge variable for the type of problem-solving represented in this study. This served to intuitively validate what has been documented in the literature. In these discussions of technical knowledge, there appeared to be some tendency for expert technicians to underestimate or "take for granted" their level of technical knowledge. In some cases, technicians failed to recognize the importance of their knowledge because it had become a rather natural part of their problem-solving activity. When encouraged to reflect on the problem-solving process, there was a tendency not to recognize the extent of their knowledge base as well as their ability to apply it to practical situations.
Practical vs. theoretical knowledge. The third theme that emerged out of the discussions had to do with a distinction between practical and theoretical knowledge. While the literature suggests that near transfer problem-solving is a blend of theoretical and applied knowledge and skills, the technicians viewed their job responsibilities as practical in nature. For example, one technician stated, "We see lots of engineers [and technicians] who have theoretical knowledge, but little ability to apply what they know. They seem to talk themselves out of making a decision...they are unsure if they are right and are unwilling to test it." While the clear preference for practical, applied knowledge was not surprising, given the background of the participants and the nature of this study, it does suggest that some additional thought should be taken in future studies to distinguishing between the various types of technical and content knowledge needed to solve different kinds of technological problems.
Experience and self-confidence. The last theme that ran throughout the various discussions focused on self-confidence and experience. Some felt that self-confidence is a function of experience while others observed that some technicians are "probably more confident than they should be because they are not yet aware of what they don't know." Others observed that a lack of self-confidence represents a barrier to innovation. One technician commented that "the second shift does the same repetitive motion as the first...many of them seem to be afraid to take any innovative actions on their own!" In summary, there was a rather consistent recognition of the importance of self-confidence in solving technological problems. What was less clear was the extent to which the variable is important or how the mechanism functions as a part of the problem-solving process.
Discussion and Implications
A primary conclusion of this study concerned the transferability and applicability of many characteristics common to general problem-solving to near transfer technological troubleshooting problem-solving. For example, Sternberg (1986) and Stevenson (1994) concluded that cognitive knowledge is an integral component of general problem-solving. Based on the results of this study, this appears to be the case with near transfer technological problem-solving as well. Another example can be found within the work of Gobet and Simon (1996) and Holyoak (1995) who concluded that a key component of effective general problem-solving is familiarity with the problem setting. Once again, based on the strong correlation value of the "years of experience" variable, this tenet also appears to apply to near transfer technological problem-solving. Familiarity is typically a function of experience or exposure to a problem setting. The amount of exposure to a problem setting can also be logically linked to an aggregate cognitive knowledge base. These conclusions are particularly important for the specific subset of technological troubleshooting problem-solving examined in this study (i.e., near transfer) and supports the conclusion that many of the same personal/social problem-solving dynamics extend to technological problem-solving (at least to the troubleshooting subset).
Experience and Knowledge
The conclusion that years of experience contributes to problem-solving expertise makes intuitive sense (i.e., information gathering and manipulation, hypotheses generation, etc.). Therefore, it was reasonable to expect that "years of experience" would emerge as a strong indicator of near transfer technological problem-solving skills. This study's findings validated this expectation and concluded that the years of experience variable was the strongest indicator of near transfer technological problem-solving competence.
The study's findings also reinforce the notion that, in order to successfully solve near transfer technological troubleshooting problems, problem-solvers must possess a firm grasp of cognitive technical knowledge. This conclusion is corroborated by the general problem investigative work of Sternberg (1986) and Stevenson (1994). The cognitive technical knowledge variable ranked second only to "years of experience" in correlation strength. The inference is clear. Employers should provide technicians with the opportunity to develop, enhance, and continually update cognitive content knowledge, particularly in areas where the technology is changing rapidly. It is also clear that the base of cognitive knowledge can be expanded and made more practical through years of experience. Thus, these two findings appear to be interrelated. Interestingly, subsequent conversations with the participants indicated this was not always the case (i.e., years of experience does not always translate into increased troubleshooting expertise).
Refined critical thinking skills has emerged as a key concern for education and industry in recent years. Several studies have concluded that critical thinking skills are germane to most effective general problem-solving skills (Pears, 1995; Polya 1957; Skinner, 1995; Watson & Glaser, 1964). Based on findings of these studies, critical thinking skills appear to involve a process where potential solutions are evaluated in an iterative fashion until a decision is made for the most useful and practical problem-solving strategy.
The findings of this study reinforce the notion that the components of critical thinking skills identified by Glaser (1941) (i.e., recognition of problems, gathering of pertinent information, recognition of unstated assumptions and values, comprehension and use of language with accuracy, clarity and discrimination) are relevant indicators of near transfer trouble shooting technological problem-solving skills. Thus, the development and enhancement of critical thinking skills is important in order to refine effective near transfer technological problem-solving skills.
One of the more important outcomes of the study involves the synergism among the top three ranked predictor variables (i.e., years of experience, cognitive technical knowledge, and critical thinking skills). When analyzed separately, each of these variables were predictive of near transfer technological problem-solving ability. When the critical thinking skill variable was injected into the prediction equation, the predictive strength of the model increased substantially. This finding suggests that while all three factors are important indicators of near transfer technological problem-solving capabilities, the development and use of critical thinking skills in the workplace is particularly important.
Previous research indicates that cognitive styles (e.g., field dependence/independence) are influenced by cognitive functions such as perception, memory, thinking, and problem- solving (Swinnen, Vandenberghe, & Van Assche, 1986). There also appears to be little indication that one cognitive style is superior to the other in general problem-solving. However, in a highly technical troubleshooting situation, it could have been anticipated that those with a field independent style would be more successful. This study's findings did not support this assumption (e.g., the superiority of field independence in technological problem-solving). The implications of this finding are important. For example, if the findings had suggested a strong preference for field independence, this would have indicated that only certain individuals are capable psychologically of being successful with troubleshooting technological problems. However, this study's results suggest that individuals possessing a range of cognitive style preferences can solve these types of problems. What remains to be studied in more depth are the mechanisms various personality types use to navigate the problem-solving process. While it is clear that field dependent and independent oriented problem-solvers approach problematic situations differently, it is encouraging to note that they may be equally well-suited for near troubleshooting transfer technological problem-solving.
Based on this study's results, the factors that influence general problem-solving ability (identified in the literature) also apply to the near transfer, troubleshooting subset of technological problems. Specifically, a cognitive knowledge base is necessary and experts display patterns of behavior that are distinct from novices. What remains to be examined is how these mechanisms work across the broader spectrum of technological problems (e.g., design and ill-structured technological problems). Considerable work also remains to better understand the factors that influence expertise. As we better understand these factors, strategies can be developed to build an appropriate base of expertise in those graduating from our educational programs.
Custer, R. L. (1995). Examining the dimensions of technology. International Journal of Technology and Design Education, 5, 219-244.
D'Zurilla, T., & Maydeu-Olivares, A. (1995). Conceptual and methodological issues in social problem-solving assessment. Behavior Therapy, 26(3), 409-432.
D'Zurilla, T. J., & Nezu, A. M. (1990). Development and preliminary evaluation of the Social Problem-solving Inventory. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 2, 156-163.
Drucker, P. F. (1993). Post-capitalist society. Harper Business: New York.
Dyrenfurth, M. J., Custer, R. L., Loepp, F. L., Barnes, J. L. , Iley, J. L., & Boyt, D. (1993). A model for assessing the extent of transition to technology education. Journal of Industrial Teacher Education, 31(1), 57-83.
Glaser, E. M. (1941). An experiment in the development of critical thinking. New York: Teachers College, Columbia University.
Gobet, F., & Simon, H. A. (1996). The roles of recognition processes and look-ahead search in time-consuming expert problem-solving: Evidence from grand master-level chess. Psychological Science, 7(1), 52-55.
Holyoak, K. J. (1995). Thinking: An invitation to cognitive science, 3, (2nd ed.), 267-296.
Jereb, J. (1996). The technical problem and its didactic function. Reports/Descriptive. (ERIC Document Reproduction Service No. ED 391 699)
Johnson, S. D., & Thomas, R. G. (1994). Implications of cognitive science for instructional design in technology education. Journal of Technological Studies, 20(1), 33-44.
Lavoie, D. R. (1991). 1991 The construction and application of a cognitive-network model of prediction problem solving. Paper presented at the Annual Meeting of the National Association for Research in Science Teaching. Lake Geneva, WI.
Newell, A., & Simon, H. A. (1972). Human problem-solving. Englewood Cliffs, NJ: Prentice Hall.
Pears, G. ( 1995). Teacher competencies in teaching thinking. Unpublished Dissertation, University of Texas.
Polya, G. (1957). How to solve it. Garden City, NY: Doubleday/Anchor.
Simon, H. A. (1996). The roles of recognition processes and look-ahead search in time-consuming expert problem solving: Evidence from grand master-level chess. Science, 7(1), 52-55.
Skinner, R. (1995). Authentic assessment: Projects for the future. Speeches/Conference Papers. (ERIC Document Reproduction Service No. ED 392 611)
Sternberg, R. J. (1986). Critical thinking: Its nature, measurement, and improvement. Viewpoints, Information Analysis. (ERIC Document Reproduction Service No. ED 272 882)
Stevenson, J. (1994). Vocational expertise. In J. Stevenson, S. Billet, G. Evans, C. McKavanagh, H. Middleton, & F. Beven (Eds.). Cognition at work: The development of vocational expertise, National Centre for Vocational Education (pp. 7-30). Research Ltd., South Australia.
Swinnen, S., Vandenberghe, J., & Van Assche, E. (1986). Role of cognitive style constructs field dependence-independence and reflection-impulsively in skill acquisition. Journal of Sport Psychology, 8(1), 51-69.
Waetjen, W. B. (1989). Technological problem-solving: A proposal. International Technology Education Association. Reston, VA.
Watson, G., & Glaser, E. (1964). Watson-Glaser critical thinking appraisal. Chicago: Harcourt, Brace and World.
Winek, G., & Borchers, R. (1993). Technological problem-solving demonstrated. The Technology Teacher, 52(5), 23-25.