Problems and Problem Attention in the Construction Sector – Understanding the Influence of Human Factors

Problems concerning quality and productivity in the construction sector have been a recurrent issue for many years and seem to remain in spite of various initiatives for resolving them. This situation is a result of human action. From social sciences we know that psychological factors crucially influence action design. Knowledge of this influence seems however to be underestimated in the construction sector, and could represent a missing link between strategies, plans and instructions, and the actions carried out. In order to prospect for new problem solving approaches we undertook a questionnaire-based survey to investigate how individuals in the sector perceive the importance and occurrence of, and attention directed to, different proposed causes of inadequate performance. The design of the questionnaire enabled comparisons of different answers to look beyond the respondents’ overt answers. The result suggests that (1) the whole problem solving situation, including individual, relational and contextual problem components should be addressed as ‘the problem’; (2) the workforce has the ambition and courage to do what is expected but does not always have adequate information and the ability or resources to do it; (3) the approach taken in this study appears to have the potential for looking behind the façade, by detecting different kinds of contradictions.


Introduction
The construction sector has for long been criticised for deficiencies and defects, cost overruns, delays, unethical business methods and the like.Despite initiatives, academic as well as industrial, to rectify the problems, a Swedish government report (Statskontoret 2009) established that: "[the] number of construction defects […] has probably not diminished.[…] Independent judges with close experience from the sector claim that the range of defects has instead increased (p.14)." In their search for explanations as to why the problems remain, some researchers have come to wonder whether or not there is something missing in today's theoretical understanding.On an overall level, Garel (2013) notes that 'while there is a global conception of [project management], there is no unified theory', but rather 'an articulated collection of best practice.' Lalonde et al. (2012) go as far as talking about a knowledge crisis within the field of project management, which 'calls for not only a re-examination of the complex relationships between project management theory and practice, but also a rethinking of project management research itself in order to better grasp these relationships.'Some critics argue that traditional research usually assumes an idealised image of the organisation as an orderly, limited phenomenon (Green & May 2005).An alternative, more fruitful approach would be to see the organisation as 'a fluid and dynamic web of complex social, economic and political relations' where 'fragmentation and conflict, are the "natural" state of affairs' (Bresnen et al. 2005).Bresnen et al. (2001) and Ofori & Toor (2009) point out that construction management research has taken a different path than mainstream organisational management research, by being more influenced by engineering, production and operations management.When construction management research has applied ideas If the area is divided into two dimensions, one representing the type of question asked - (A) what happens (and how can it be improved)?and (B) why is it happening (and how can it be improved)?, with the other dimension representing the object of the study -(C) effects of human action and (D) human action per se, it can be seen that very few studies are concerned with the combination BD, why is it happening and human action per se.The common approach appears to be the combination AC, what happens and effects of human action, where the human individual or the working team is viewed as a 'black box'.
Even if not in abundance, there are recent examples of research trying to look into the black box and connect inner human functions to what is happening in construction.One example is Sunindijo & Zou (2013) who investigate the role that emotional intelligence might have on construction safety.Lowe & Skitmore (2011) investigated how the learning climate in some practices correlates to individual learning styles.Chiu and Ng (2013) link work group identification to organisational commitment and outcome.Boyd (2013) makes an alternative interpretation of the purpose of knowledge management, from 'to think better about practice' to 'supporting people to act better in practice.'

Human Action and Information Distortion
Human action is designed in interplay between the actual situation and the acting individual.Correct information is considered to be of central importance in order to master the complexity of construction (Winch 2002).At the same time, we know from different disciplines of social science and from experiences in the domain of practical project management (Kutsch et al. 2011;Brady & Maylor 2010) that humans regularly distort information for different reasons.Russo et al. (2008) go as far to say that '[the] distortion of information during a decision seems to be nearly ubiquitous.'This distortion is normally blind to the individual, which has been accounted for by several authors, e.g.Argyris et al. (1985), Khaneman (2011), Cohen & Sherman (2014).In psychodynamics theory, information distortion is explained to be a result of psychological defence mechanisms, serving to help  and Building, 14 (2) 1-17  3 the individual avoid anxiety.These mechanisms operate automatically and beyond our awareness in situations that are at risk of becoming embarrassing or threatening (McWilliams 2011;Cramer 2000).Social psychologists have noted that people regularly make attribution errors when trying to create a full picture from the fragmented information available (Heider 1958;Malle 2008).Argyris et al. (1985) have recognised a gap under certain conditions, between the values that persons express (espoused-theory) and the values implied by their action (theory-in-use).The reason is said to be an inconsistent and contradictory strategy or thinking pattern.
In their review of a critical perspective of project management theory, Bresnen et al. (2005) have noticed similar inconsistency in the 'quite significant gaps that can arise between the rhetoric and reality of management initiatives such as lean construction, business process re-engineering and partnering'.Another example of this gap is the stark finding reported by the investigators in an official inquiry into Swedish construction sector (Byggkommissionen 2002): "[a] lasting impression of the commission's work is that the actors [in the construction sector] generally agree about the problems presented, but consistently do not consider themselves to have anything to do with either the genesis of the problems or their solutions".
This attitude, if it is prevailing in the construction sector, indicates an inconsistency in the actors' perceptions of the problems in the sector that might contribute to an explanation the persistence of the problems.
The following sections discuss three central themes of the study: (1) the situation, (2) the individual and (3) the individual's interpretation of the situation.

The Situation, the Individual and the Interpretation The Situation -Joint Problem Solving
Problem solving includes on the one side, the trouble or inconvenience to be eliminated, and on the other side difficulties of the problem handling process itself.For example, Oyedele (2013) identifies three categories of demotivating factors influencing architects in their daily work: project related, design team/co-worker related and organisational criteria.Together, these problem components constitute the task to be solved.
In order to identify such additional process-related problems, a joint problem solving situation is visualised in Figure 1 (a).In the figure, the problem solvers, pulled out of their context, are provided with boundary conditions, i.e. output and input, according to systems theory.Several problem components can be identified:  The trouble/inconvenience that should be removed (P T ).
 The problem solvers themselves (P I ).If the wrong actions are carried out the problem solving process will fail or at least be inefficient. The relations between the problem solvers (P R ).Emerging phenomena on the group level can trigger reactions in the individual, e.g.social influence and group mechanisms as power, harassment, stereotyping (Hamilton, 1981). The problem solving capacity of the group as a whole and the single group members will be influenced in different ways by the groups' context (P C ). Mental influences are e.g.demands, norms, culture and power relations.Physical influences can be exerted by weather, failing deliveries, accidents etc.  Side-effects (P E ) of the problem solving process which can be negative or positive. Dynamic, i.e. time-related, problems due to change and development (P D ).Ongoing processes will gradually change the problem and the conditions surrounding them.

The Individual -Causes for Inadequate Performance
When looking for explanations to the problems in the construction sector the starting point has been four explanations a lay person often uses when attributing causes to other people's failures.The four chosen categories chosen are: they a) did not understand how to (NU); b) did not want to (NW); c) did not dare to (ND) or; d) were not able to (NA) solve the formulated problem.
NU: This category refers to an inadequate or incomplete mental model in a specific moment (Johnson-Laird 2004).It can be inadequate or incomplete in different ways.Here, the information distorting mental mechanisms discussed earlier play an important role.
NW: This category relates to different theories of motivation and drivers (e.g.Pritchard and Payne 2003;Deci & Ryan 2002;Oyedele 2010Oyedele , 2013)).People have innumerable different actions to choose from in a specific moment, and there are reasons why a specific choice is made.Both logical thinking and emotions are involved.
ND: One reason for not doing the 'right thing' is fear.Fear is an important mechanism that protects people from doing things that can be fatal in one way or another (Dozier 2000).
There is instant fear appearing in the situation, there is anxiety that can be explained as fear for something happening elsewhere or in the future, and there is the fear of experiencing abhorrence or losing control (McWilliams 2011).
NA: Let us say that a person understands how to do the 'right thing', wants to do it, and dares to do it, but still it is not happening, then the person is not able to do it.Being able to do something depends on a person's knowledge, practical skills and available resources in a specific situation.In this category, knowledge stands for divided and delimited chunks of operable knowledge within different areas, whereas more integrative strategic knowledge or insights belong to the first category related to understanding.
Other reasons: For our purpose, the four above categories have been considered to comprise all human causes to inadequate performance.This means that other causes, as for instance 'I am not allowed' or 'I am too tired' etc., should be placed under the four categories.For example, the first can be placed under 'do not dare' implying that the individual has a choice.The second can be placed under 'do not want to' or 'is not able to', depending on how exhausted the person is.

The Interpretation and Information Distortion
The 'lasting impression' of the investigators quoted earlier (Byggkommissionen 2002), indicates some form of information distortion.There is agreement as to the existence of information distortion among different psychological traditions.However, they have different explanations of the underlying causes and, as a consequence, also how to deal with their negative effects.For example, the psychodynamic tradition focuses on unconscious psychological defence mechanisms (McWilliams 2011), cognitive psychologists point at people's limited information processing capacity (Reason 2007), while the sociopsychological tradition refers to techniques such as social attribution (Malle 2004, Barazova & Hancock 2012), and social stereotyping (Hamilton 1981).Psychological defence mechanisms are designed to help a person escape from anxiety that otherwise could emerge in embarrassing or threatening situations.Social attribution means that people ascribe identity and traits to other people (even to themselves) and reasons for their behaviour.However, attribution comes with a number of attribution errors, e.g. the fundamental attribution error (Ross 1977) and the superiority bias (Hoorens 1993).The fundamental attribution error is a tendency for observers to overestimate the causal influence of personality factors on behaviour and to underestimate the causal role of situational influences.The superiority bias, also known as the "better-than-average-effect", is a tendency to overestimate one's own positive qualities in relation to others', and underestimate the negative.The gap between 'espoused-theory' and 'theory-in-use', earlier mentioned, is also associated with what Argyris et al. (1985) call a kind of 'blindness' that makes the individual unable to detect the gap.Kahneman (2011) reports on a similar kind of 'blindness', where he describes the human mind as consisting of two parallel thinking systems.System 1 is fast, intuitive and energy-efficient; it is rather accurate most of the time but sometimes makes serious misinterpretations.System 2, on the other hand, representing effortful thinking, is slow, deliberate and energy-consuming.Kahneman has shown that people often are reluctant to engage system 2, because of its strenuous character.

Research Method Questionnaire -based Survey
Since we were trying to find explanations for the situation within a whole industrial sector, we had to assure ourselves that the results of our investigation not only reflected local conditions.That means that we needed to reach out to different parts of the country, different types of organisations, different types of individuals etc.Since we were also looking for ideas that could explain why people continually are designing and carrying out actions that will lead to faults and criticism in a later stage, we needed to address the single individuals who carry out the actions.According to Bryman (2001), a questionnaire survey best meets the above requirements.Below, the design of the questionnaire and the principles for data analysis, the sample and the manner of distribution are presented.
The aim of the questionnaire was to investigate: 1. the problem apprehension of actors in the construction sector; 2. the actors' problem attention; 3. the specific individual or situational factor -if anythat can explain the apprehended problems; and 4. the presence of information distortion.

Design of the Questionnaire
In order to be able to study separately, in more detail, the influence of the different situational and individual factors that determine the conditions for human action design, a question generator was developed.This was done by combining the two models earlier presented, i.e. the situational problem components (P T , P I , P R , P C , P D and P E ) and the causes for inadequate performance (NU, NW, ND, NA) into a 24-frame matrix (see Figure 2).and Building, 14 (2) 1-17 6 The 24 phenomena generated in this way are presented in Table 1.

Table 1 The 24 phenomena, each representing a unique combination of the problem-solving situation and reasons why individuals do not act in an anticipated rational way
The 24 phenomena were then used to generate questions concerning four specific aspects of each phenomenon: 1) its importance, 2) its occurrence, 3) the attention paid to it in daily work in the construction sector; and 4) the attention paid to it in daily work by the individual respondent; in all 4*24=96 questions (see Figure 2).Thus, the four questions concerning the first phenomenon read: Importance Since we are looking for contradictions and inconsistencies, one issue in the design is to prevent adjustment or 'balancing' of previously answered questions.This was achieved by presenting the questions, one by one, in a specific order.

Analysis of the Results
The questions were answered on a five-point Likert-scale with the answer alternatives: 1. No/Very small; 2. Small; 3. Moderate; 4. Large; 5. Very Large.Because every question had to be answered to enable the fixed presentation order, the Likert-scale was extended with three administrative answer options: 6.I do not know; 7. I do not understand (the question); 8.The question is not relevant.The responses were processed and analysed in the following ways: I.A frequency analysis was performed to get information about the distribution of the different answers regarding importance, occurrence, sector attention and individual attention.II.A relevance index was calculated for each phenomenon by multiplying the importance of a phenomenon with its occurrence, i.e. by multiplying the answer value (1-5) for the two aspects.The rank order of each of the investigated phenomena with respect to its arithmetic mean of the relevance index (n=valid answers) was calculated: III.In order to look for information distortion, each respondent's answers regarding the sector's attention and the individual's own attention concerning the 24 statements were compared.
IV. From a rational point of view it could be expected that high attention is directed to phenomena of high relevance and low attention to phenomena of low relevance, from both individuals and from the sector.Eight cases of different combinations of relevance, sector attention and individual attention are possible (see Table 2).Case 1 and 8 represents what can be considered as rational behaviour, while the rest indicate some kind of mismatch.A Pearson Chi-square 1 analysis was used in order to analyse the difference between observed and expected frequencies 2 for each phenomenon and case.Since the cases represent different mismatches we are interested in finding those that more people than expected experience.Note that we are not looking for the chi-square value per se.
Table 2 Eight possible combinatorial consequences of the respondents' opinions of the conditions in the construction sector.
V. Another way of understanding the results, looking behind the respondents' overt answers, is to bring them back to the two dimensions of each aspect of the question generator in Figure 2, i.e. situational problem component and cause for inadequate performance, to see if any pattern emerges.Does any individual cause or situational component stand out compared to the rest?This is done by completing the ranking tables of Importance, Occurrence, Relevance, Sector attention and Individual attention (in Table 3) with their coding for situational problem component and cause for inadequate performance.

Sample
The questionnaire was distributed to a representative sample of white-collar personnel working in different positions in the construction sector in Sweden.The membership lists of three influential 'white-collar' branch organisations were used: a) the Swedish Society of Civil and Structural Engineers, SVR (612 respondents); b) the Swedish Association of Construction Engineers, SBR (554 respondents); and c) the Swedish Association of Architects, SA (566 respondents).
Together, the three organisations have about 15,000 members (SBR 2700; SVR 4000; SA 11,500 of which only the categories SAR/MSA and A MSA, relating to building architects, were used) which roughly correspond to about 10% of all architects and engineers active within the construction sector.However, there are differences between the organisations.SA is a trade union and has enrolled about 90% of all architects, while SBR and SVR are more interest-based and organise a much smaller part of the total number of engineers.A random number generator was used to select page numbers in the membership lists.Each person on a selected page having an e-mail address and, in the case of architects, the right categorisation was included in the sample.

Distribution Method
Inquiries were sent via email with instructions on how to login and answer the questionnaire.This invitation was followed-up by three reminders, with one week's interval, to those who had not returned the completed questionnaire after two weeks.

Response Frequency
In all, 304 responses were received.Due to the low response rates (~20%) a dropout analysis was performed on 284 randomly-selected people who had not answered the questionnaire and who were then contacted by telephone.It was found that 24.3% of the respondents did not receive the login instructions due to invalid e-mail address, 2.5% were not available at the time for the survey, and 7.7% could not be traced at all whereas 18.7% did not work in the construction sector, and therefore did not belong to the target group.Furthermore, 33.8% had received the questionnaire and found it interesting but could not allot the time needed to answer it, 6.0% were not interested in participating and 1.8% had lost their login information.The remaining 5.2% referred to the questionnaire layout, personal values, insufficient secrecy routines or technical problems as reasons for not answering the questionnaire.A re-estimation of the response rate based on the results of the dropout analysis indicated that the 304 answers correspond to ~52% of the 'valid' respondents, which corresponds to 'barely acceptable' in Mangione's (1995) categorisation.

Ranking of the 24 Phenomena
As earlier explained, the questionnaire was used to obtain two kinds of result, the respondent's direct answers and combinations or comparisons of the direct answers.A frequency analysis was performed to get information about the distribution of the different answers.The result is presented in Table 3, in the form of ranking tables.A low number indicates a high ranking, i.e. an important or frequently occurring phenomenon or phenomenon attracting high attention.
Table 3 The 24 phenomena ranked with respect to the four aspects and combinations or comparisons of them.
* The 'Importance' of phenomenon 20, marked with an asterisk is considered flawed since 145 out of 304 respondents (43.2%) marked answer alternative 7. I do not understand (the question).This indicates that this particular question was inadequately formulated and should therefore be handled with caution.The combined answers regarding relevance, case 1 and case 2, marked with question marks should also be considered as flawed since they build on the same answer.
In Table 3  attention and individual attention.The second cluster, "Combined Answers," shows the result of the comparisons of the answers processed by the authors according to the earlier described analyses II, III and IV.
In a practical context the most highly-ranked phenomena are the most salient and probably also the ones with the highest potential to make a difference.To make it easier to distinguish those, the six (25%) most highly-ranked phenomena in each aspect of the investigation have been framed with a thick-line.

Respondents' Direct Answers
The rankings in the 'Direct Answers'-columns of Table 3  Sector attention: The six phenomena of highest perceived sector attention are (in ranking order): 3. the case that someone experiences a risk to be injured; 17. insufficient information about changed conditions; 1. unclear target/problem formulation; 12. insufficient/inefficient communication in the working team; 4. lack of skills in the working team; 13. misunderstanding of external working prerequisites/conditions.

Individual attention:
The six phenomena of highest perceived individual attention are (in ranking order): 1. unclear target/problem formulation; 17. insufficient information about changed conditions; 12. insufficient/inefficient communication in the working team; 13. misunderstanding of external working prerequisites/conditions; 21. insufficient/lack of a comprehensive view; 3. the case that someone experiences a risk to be injured.

Combined Answers
Relevance: A relevance index was calculated for each phenomenon by multiplying the importance of a phenomenon with its occurrence, i.e. by multiplying the answer value (1-5) for the two aspects.The ranking can be seen in Table 3.The six most relevant phenomena are: 1. unclear target/problem formulation; 17. insufficient information about changed conditions; 12. insufficient/inefficient communication in the working team; 21. insufficient/lack of a comprehensive view; 9. the case where people disagree in the team; 16. too many or conflicting demands.

Sector and individual attention:
In order to look for information distortion, each respondent's answers regarding the sector's attention and the individual's own attention concerning the 24 statements were compared.Three kinds of result are possible: (1) the respondent finds the sector more attentive; (2) the respondent finds himself/herself more attentive; and (3); the respondent finds himself/herself and the sector equally attentive.The outcome is shown in Table 4.The table header shows the ranking.The first row shows the phenomena in ranking order.The second row shows the number of respondents who perceive the sector paying more attention to the particular phenomenon than they are themselves.The third row shows, conversely, the number of respondents finding themselves being more attentive than the sector.The fourth row shows ∆ = (row 3 -row 2), which also determines the ranking.
Table 4 Comparison between the responses to aspect 3: about the sector's attention and the responses to aspect 4: about the individual respondent's attention.All aspects can be seen in Figure 2.
Since the sum of the individuals equals the sector there should ideally be no gap.Yet, in every comparison ∆ is positive, meaning that the individuals are perceiving themselves to be more attentive than the sector.The six phenomena earlier considered as the most relevant have been marked with thick lines in the first row.The six phenomena showing the largest difference (∆) are: 1. unclear target/problem formulation; 21. insufficient/lack of a comprehensive view; 9. the case where people disagree in the team; 17. insufficient information about changed conditions; 8. the case where it is difficult to prioritize among different tasks; 13. misunderstanding of external working prerequisites/conditions.

Comparison between relevance and attention:
In order to see if due attention is directed towards the most relevant phenomena, the deviation between observed and expected frequencies for the six (2-7) mismatching cases in Table 2 was calculated using a Pearson Chi-square analysis.The result is shown in Table 5.
Case 2 represents phenomena that a respondent estimates to be of high relevance and to which he/she at the same time considers him/herself to direct a high degree of attention while the sector is considered to direct a low degree of attention.The six phenomena with the largest deviation (in the black square) are: 21. insufficient/lack of a comprehensive view; 14. cultural barriers/considerations; 4. lack of skills in the working team; 1. unclear target/problem formulation; 17. insufficient information about changed conditions; 12. insufficient/inefficient communication in the working team.
Case 5 represents phenomena of estimated low relevance where both the individual and the sector direct a high degree of attention.The six phenomena with the largest deviation (in the black square) are: 16. too many or conflicting demands; 6. the case where a person is not interested in the problem being solved; 4. lack of skills in the working team; 1. unclear target/problem formulation; 3. the case that someone experiences a risk to be injured; 17. insufficient information about changed conditions.Some of the phenomena can be found high in both case 2 and case 5 (1, 4 and 17) which indicate that the sector is divided into different camps.

Pattern regarding cause and problem component:
In order to see if any particular situational problem component or individual cause for inadequate behaviour stands out, according to analysis V, the ranked lists of Importance, Occurrence, Relevance and Attention, in Table 6, has been completed with their coding for "cause for inadequate behaviour" (NU, NW, ND and NA) and "situational problem component" (P T , P I , P R , P C , P D , P E ).To make it easier to distinguish any pattern, the initial letter has been removed in the table, i.e.NU=U, PT=T and so on.
Table 6 Coding regarding cause for inadequate behaviour and situational problem component.
In order to make it easier to discern any pattern just the distinguishing letter for cause and problem component is used so that e.g.U=NU=do not understand and T=PT=trouble/inconvenience.

Discussion Results
We asked white-collar personnel in the Swedish construction sector about their views on certain problems and conditions in their daily work.The question about how practitioners in the Swedish construction sector assess different problems and conditions in their daily work, is answered by the questionnaire and presented in Table 3.Among the six most relevant problematic phenomena, i.e. phenomena that are both important and frequent, three can be related to leadership outside the working team, i.e. 1. unclear target or problem formulation; 17. insufficient information about changed conditions and; 16. too many or conflicting demands.One can be related to both leadership and the working team, i.e. 21. insufficient or lacking comprehensive view.The remaining two can be related to the working team i.e., 12. insufficient communication in the working team and; 9. the case where people disagree in the team.
The first, and most salient of all problem components, 1. unclear target or problem formulation, indicates a missing link between rhetoric and action.A clear vision, goal or objective is central in several problem solving or change methods (Kanter et al. 1992;Kotter 1996;Luecke 2003).According to the respondents, this is obviously not the case in practice.From the answers in the questionnaires, it is not however possible to conclude why the target or problem formulation is regarded as unclear.This has to be further investigated.
The presence of patterns in the way practitioners in the construction sector perceive their situation and that can explain certain problems, may to some extent be revealed by the questionnaire.There are some contradictions in the material that might indicate inconsistencies.The most obvious comes from the comparison of the single individual's appraisal of his/her own attention to important issues and his/her appraisal of the whole sector's attention to the same issue.In all 24 phenomena, there are more individuals thinking that they are more attentive than the sector as a whole.Since individuals constitute the sector the answers should be in balance.Although the result, per se, is not too surprising, given well-known human traits such as e.g. the superiority bias (Hoorens 1993), there is still a risk that this misleads the conception of reality and hinders exchange of correct information.The difference is especially clear when it comes to the most relevant phenomena, where the six most relevant are situated among the nine showing the highest difference in attention.This result lends some support and explanation to the Byggkommissionen (2002) investigators' impression that individuals in the sector neither see themselves as part of, nor the solution to, the sector's problems.If this is true, it alone could explain the much criticised inertia in the sector.
Furthermore, do the answers to the questionnaire reveal if some prerequisite component is missing?When the answers are brought back to the situational dimension in the question generator it appears that the technical component (P T ) and the relational component (P R ) are considered the most relevant and are the two most noted by both the sector and the individuals.However, the answers are quite evenly distributed among all the six components, which support the use of the extended problem formulation in Figure 1 (b).
Do the answers to the questionnaire reveal if psychological factors, e.g.information distortion, might influence the way problems are perceived?When brought back to the dimension of causes of inadequate performance, there is a more salient division into two groups, where 'Do not understand' and 'Are not able to' dominate in relevance as well as in both sector and individual attention.'Do not dare to' and 'Do not want to' are constantly ranked low suggesting that the respondents do not consider those aspects as particularly relevant and therefore do not pay much attention to them.This indicates that additional efforts should not be spent on merely motivational events, but rather on information, communication, education and enhancing equipment that can raise the individuals' performance capacity, which in itself is motivating.
When comparing relevance and attention, most cases seem to be in accord.There are, however, exceptions in case 2, i.e. issues of high relevance to which the sector is considered to pay little attention while individuals pay high attention, and in case 5, i.e. issues of low relevance to which both the sector and the individuals pay high attention.A situation, where the individual experiences that 'the others' do not direct proper attention to a relevant phenomenon, or that both the sector and individuals are directing high attention to phenomena that are considered to be of low relevance, might provoke tensions, e.g.frustration, apathy or mistrust, which could influence communication and cooperation negatively.

Limitations
Even though questionnaire-based surveys are a well-established form of inquiry, two features here can be regarded as somewhat unorthodoxthe question generator and the combinatory procedure.The question generator was of significant help in formulating each question.This strict question formulation procedure meant that answers could be coded in order to detect underlying patterns in cause or situation.
It should be noted though, that the conclusions made about individual causes for inadequate performance and situational problem components results lean heavily on the correlation between one single question and the referred phenomenon.Here, it was considered to be sufficient with one question because the aim was not to find some evidence-based 'truth' but rather probe for new ideas as input into our learning cycle and method development.The ranking of the phenomena per se, and the contradiction between sector attention and individual attention, is more straightforward and therefore not associated with such interpretative risk.
What is perhaps the most interesting aspect of the combinatory procedure is the chance to look behind the answers given by the respondents by combining them in different ways.In this way, inconsistencies or contradictions in the respondents' view of reality can be identified.However, it can be difficult to define the combinatorial consequences in a perfectly correct manner, which is why the results should be handled with both care and caution.But if this is done properly, it can generate outcomes that can be hard to achieve otherwise.
In this investigation the random sample represents a cross section of the construction sector.However, it does not tell if there are local variations.For example, newcomers may see the situation differently than more experienced colleagues, client side representatives differently than entrepreneurs and managers of large projects differently from managers of small projects.

Conclusions
One central idea in this investigation has been that it is individuals' actions that lead to the results that are so often criticised.We have assumed that an individual's actions are conclusively influenced by his/her inner mental constructs and appreciation of the situation at any given time.This appreciation is not a hard copy of reality, but rather a personal interpretation based on memories, fragmented perceptions and inner aptitude.The investigation has shed some light on what individuals think about phenomena in their daily practice and which probably will influence their action design, consciously or unconsciously.
First, the result shows that it is necessary that problem definitions are widened to include the problem solving situation per se with its combination of individual, relational, contextual and dynamic factors.This is easy to overlook, but is essential when designing measures that should lead to intended results.Such an expanded problem view is however at risk of becoming limitless from a comprehensive theoretical perspective and is, thus, impossible to manage in a practical situation.In such a practical problem solving situation, relevant factors only should be handled.Here, the individual-situation matrix, or question generator, see Figure 2, can be of help in methodically finding relevant factors to a particular problem solving situation.
Second, the emerging picture shows that the workforce has the ambition and courage to do what is expected but does not always have adequate information and the ability or resources to do it.However, the result also indicates the existence of some kind of information distortion that might affect the way they understand the world.
Third, the approach taken in this study appears to have the potential for looking behind the façade, by detecting different kinds of contradictions.Investigations that attempt to understand how people design their actions and why they do not perform as anticipated, starting in their daily routine, could provide valuable complementary information as a basis for performance development and change in the sector.
Sunding, L and Ekholm A (2014) 'Problems and problem attention in the construction sector -understanding the influence of human factors', Australasian Journal of Construction Economics

Figure 1
Figure1(b) expresses the relation between the task to solve, P T ', and the trouble or inconvenience, P T .While P T was the reason to enter into a problem solving activity in the first place, P T ' is the problem or task that has to be solved in order to get rid of the inconvenience.

Figure 1
Figure 1 a) The problem-solving situation.B) The problem component 'formula'.
Sunding, L and Ekholm A (2014) 'Problems and problem attention in the construction sector -understanding the influence of human factors', Australasian Journal of Construction Economics

Figure 2
Figure 2 The question generator, relating situational problem components to causes for inadequate performance, including the four aspects (importance, occurrence, sector attention and individual attention).
(1:1): What negative effects on effective, common value creation has: unclear target or problem formulation?Occurrence (2:1): To what extent exists, in today's construction practice: unclear target or problem formulation?Sector attention (3:1): To what extent does the construction sector preventively address: unclear target or problem formulation?Individual attention (4:1): To what extent do you yourself preventively address: unclear target or problem formulation?
are based on the respondents' given answers in regard to the four aspects of importance, occurrence, sector attention and individual attention.
Occurrence: The six phenomena of highest occurrence are (in ranking order): 21. insufficient/lack of a comprehensive view; 16. too many or conflicting demands; 1. unclear target/problem formulation; 17. insufficient information about changed conditions; 8. the case where it is difficult to prioritize among different tasks; 12. insufficient/inefficient communication in the working team.

Table 6a regarding
importance shows a pattern in cause where NU, i.e. do not understand, is generally highly ranked and ND, i.e. do not dare to, has a low overall ranking.No clear pattern can be established regarding the problem component in the same way.Table6bshows that the cause of high occurrence can be derived from both NU and NA, i.e. 'are not able to' and phenomena linked to ND and NW, i.e. 'do not want to', attract a low ranking.Concerning problem components, PI tends to have a low ranking.The relevance index in Table6ccorresponds with the tendency regarding cause in Table6b, while the problem components are evenly distributed.When it comes to sector attention, in Table6d, the tendency in cause recurs, except for a ND in the highest ranking, representing the sector's high attention on 'the case that someone experiences a risk to be injured.'Inproblemcomponent there is a clear emphasis of PT in high rank and a cluster of PE in the middle of the span but it is otherwise evenly distributed.In the last table, Table6e, individual attention, there is again repetition in cause and a non-specific pattern in problem component.