Change Order Management Factors in Building Projects in Northern Nigeria

A change is an amendment or addition with respect to the original plans, specification or other contract documents and vital factor in construction project management. A change order is an instruction from an employer approving a change. Construction projects in Nigeria have suffered deleteriously as a consequence of extensive change orders, which has manifested in cost and time overruns, disputes, arbitration, litigation and even the abandonment of projects and this negatively impacts on the efficiency of the Nigerian construction industry.This paper explore the extent change order management strategies correlate with overall building project in Nigeria. Thirty strategies for managing change order were identified through intensive literature search. Twenty -one factors peculiar to the Nigerian construction industry were used to develop a questionnaire. A total of 323 questionnaires were administered through stratified random sampling to respondents in the cities of Abuja, Kano and Bauchi. The study targeted construction Consultants namely Architects, Quantity Surveyors Building Engineers, Service Engineer (Electrical & Mechanical), Contractor and Building owners in three sub-study cities. The study recorded an overall response rate of 80.8 %. Using SPSS version 22, the questionnaire was subjected to reliability test. Statistical analysis was used using average index; spearman correlation coefficient; factor analysis and multiple regression were used to analysis data from survey questionnaire. The study reveal that change orders are better manage at the design stage and that clear and thorough project brief, better initial planning and thorough detailing of design are the three topmost management strategies. The study established that overall project success is impacted by change order management with a R of 0.252, meaning that change order management at the design stage accounts for 25.2% of the variation in building project outcomes. The Spearman's (rho) also show strong agreement among respondents. The study concluded that proper adoption and use of project implementation phase principles by project sponsor, increased effort at the design stage of projects by project participant and allocation of adequate resources to project requirement will help minimize change orders in project delivery in Northern Nigeria

responsible for cost overruns of between 25-78% and time overruns of between 27-68% of projects. As a result of the important role this industry plays in a nation's economy, it is very important to maintain successful delivery of construction project in Nigeria. Effective management of change orders to avoid failed projects could help in this regard. Although, many studies related to managing change orders have been conducted in other countries (Sun et al., 2006;Motowa et al., 2007) little or no research was found that illustrate to what extent change order management strategies contributes to building project outcome. This paper therefore explore extent to which change order management strategies correlate with overall building project success.

Objective
(i) To evaluate management strategies that can significantly curtail change order in building project; and (ii) To establish the relationship between change order management strategies and building project success.

Research Background
Change order is unavoidable in construction projects because of the multifarious nature of building projects. Oloo, Munala and Githae (2014) asserted that change orders are common to all types of projects and therefore play a fundamental role in determining final project cost and time. According to Pourrostam and Ismail (2011) the reason for failure by contractor to complete projects within budgeted time are because of change orders. Erdogan, Anumba, Bouchlaghem and Nielsen (2005) pointed out that changes will never disappear, therefore the best alternative is to manage them to prevent negative consequences. Ibbs et al., (2001) on their part concluded "that the litmus test for successful management should not be whether project was free of change orders, but rather, if change orders were resolved in a timely manner to the benefit of all the parties and the project".

Change Order Management Factors
A change order is intricate information transfer that has to be dealt with carefully, otherwise dispute related to cost and time of work may occur between client and contractor. " Change order is complex because it involves all construction team, together with a huge amount of information that either has to be ask for, sent, checked, corrected, permitted, classified conveyed or submitted, among other things" (Charoenngam, Coquinco, & Hadikusumo, 2003). Owing to this complexity, methods to manage change orders have been proposed in research studies (Ibbs, et.al., 2001;Mokhtar, Bedard, & Fazio, 2000). Table 1 shows the tabulation of 30 management strategies identified in literature. Twenty -one strategies found to be peculiar to the Nigerian Construction Industry were regrouped under two categories: Design and Construction stage. These grouping helped in developing an all-embracing enumeration of management strategies for change orders. These strategies formed the basis for the questionnaire survey (Table 3).

30
Comprehensively analyse and promptly decide using a computerized knowledge-based decision support system.

Project Success Criteria
Project success has been measured in diverse ways over the past two and half decades. However, appraisal was determined mainly by meeting three decisive factors of time, cost and quality (Atkinson, 1999). However, many researchers have disagreed with these criteria as the only way to measure outcome, since project outcomes are more complex because their characteristics vary from project to project. According to Shenhar et al., (1997) cited in Al-tmeeny, Abdul-Rahman & Haron (2010) although time, cost and quality are handy measures, they have been disapproved for not really being one homogenous measurement. These shortcomings of the conventional way of measuring success led to additional success measures proposed such as: satisfaction of interpersonal relation with project team members, client's satisfaction, fitness of purpose, profitability, and project goals (Frodell, 2008;Lim & Mohammed, 1999;Pinto & Pinto, 1991;Pocock, Hyun, Lim, & Kim, 1996). This paper considered only cost and time as measures of project success because they are the objective measures mostly referred to in the Nigerian construction industry. Other measures were included as dummy variable to ascertain this measures.These factors form part of the questionnaire for the survey as shown in Table 2. Project outcome(product,system,service etc) met all technical specified requirement 5 Environmental Impact Assessment (EIA) score 6 Projects meets stakeholders satisfaction 7 Satisfaction of interpersonal relations with project team member 8 Comprehensive Contract documentation 9 Profitability 10 Projects outcome was used for it intended purpose. 11 Projects adherence to quality based on baselines goals ,target and expectations Source: Kylindri, Blanas, Henriksen & Stoyan (2012)

Ranking of Management Strategies and Success Factors
The management strategies and success factors were analyzed and ranked according to respondents' responses based on arithmetic mean value and standard deviation. Table 3 shows the combined ranking of respondents in management strategies of change orders.  Table 3 shows the descriptive statistical analysis of the combined weighted means of the respondent, level of significant of management strategies during design stage ranges between 4.38 -3.02. This shows all identified factors are significant to controlling change orders. This finding indicates that the design stage was considered as the most important time-line for executing the most effective controls for change orders because the first five management strategies appeared under the design phase. A possible explanation for this is that changes during design phase do not require any rework or pulling down because no site work is involved. This finding aligns with those of Arain and Pheng (2007) and Motowa (2005) that change orders are better controlled at the design level. Table 4 also shows descriptive statistical analysis of respondents' views on success criteria. The outcome reveals that the two most ranked success criteria are: project completion within budget with mean value of (4.34) and project completion within time with mean value (4.23). This finding collaborate those of Atkinson (1999) and Chan (2000) that identified cost ,time and quality as key performance indicators of project success in their studies.

Level of Agreement among Respondents
To determine whether there is degree of agreement among the three groups of respondents with respect to their views of the change order management factor, Spearman rank correlation coefficient was used.  Table 5 indicated strong agreement amongst respondents in the ranking of the management strategies. This finding suggests the rejection of the null hypothesis Ho and alternative hypothesis accepted. Therefore, it can be concluded that there is a significant degree of agreement between the groups of respondent. Next the variables were assessed for their suitability of the factor analysis application. These variables were subdivided into two groups under design phase (11 variables) and construction phase (10 variables) Table 6 and 7 shows that Kaiser-Meyer-Oklin (KMO) value was 0.737 and 0.70 for design and construction item respectively, exceeding the recommended value of 0.5 (Field, 2009) an indication of factorability of the variables. The Bartlett's test of sphericity (55) = 608.257, p< 0.001 and Bartlett's test of sphericity (45) = 465.309, p< 0.001 for design and construction item respectively. This KMO value verifies the sampling adequacy for the analysis and the Bartlett's test of sphericity. The results of these confirm that the sample data were suitable for factor analysis.
Factor analysis was then carried out to examine the relation between observed and latent variable (s). The eigenvalues criterion stated that each component explained at least one variable's worth of the variability and for that reason component with eigenvalues greater than 1 should be taken. This analysis yielded three components with eigenvalues bigger than 1, accounting for 52.111% of the total variance in the 11 management strategies at the design phase and three components with eigenvalues bigger than 1, accounting for 52.313% of the total variance in the 10 management strategies at the construction phase. (Table 8 and 9).     For better interpretation of these two components, Varimax rotation was performed, suppressing loading less than 0.4 this cut-off point is appropriate for interpretative purposes (Steven, 2002cited in Field, 2009. Varimax rotation solution is present in Table 12 and 13 revealed the existence of simple structure with components showing a number of strong loadings, and variable loading substantially on two components. Award contact on fixed sum basis .741 Note. (i) Only loading above 0.4 are displayed; (ii)Value Engineering is a complex variable(i.e., an item that is in a the situation of crossloading)  Vol. 12, No. 1;2016 In order to establish the model of the relationship between management strategies and building outcome multiple regressions were performed using management at design stage and management at construction stage as independent variables with building outcome as the dependent variable. Prior to performing the multiple regression analysis a multicollinearity of independent and dependent variable was checked. Multicollinearity is the correlation between the variables in the model. According to Pallant (2001) a correlation relationship > 0.3 and 0.7 is preferred. In Table 14 the correlation values of 0.504 and 0.444 were recorded for management at design and construction stage respectively. The result of these test show that all factors were appropriate for regression analysis.
The model Summary tells us whether the model is successful in predicting building project outcome (Table 15). Since the regression was hierarchical method; each set of summary statistics was repeated for each stage in the hierarchy. In the column regarded as R are the values of the multiple correlation coefficients between predictors and the outcome. The column characterized as R 2 is a measure of how much of the changeability in the outcome is accounted for by the predictor. For the first model its value is .252 which means that change order management at the design stage accounts for 25.2% of the variation in building project outcomes. However when the other predictor (change order management at the construction stage) was included (Model 2) the value increased to .271 or 27.1% of the variation in building project outcome and it was not significant with p-value > .005. This result suggests that it is not be appropriate to include all independent variable (management at construction stage) in the regression model.    It can also be seen in the second model that p value of .007 > .005 making it not significant. This finding indicates that change order management at the construction stage should not be included as a predictor variable in the regression model.

The basic equation of multiple regressions is:
Outcome i = (model) + error i It is essentially the same for simple regression except that for every additional predictor included. In this study two predictor variables are used therefore the will be: It is important to note that since the second predictor in this study is not significant the equation becomes: Replacing the variable value as given in Table 17: Outcome = 1.13 + 0.77 ( X i1 ) + ε i This finding indicate that as change order management effort increases by one (1)

Discussion
Clear and thorough project brief is the first most important change order management strategy for building projects in northern Nigeria. The findings seems to suggest that project briefing by Nigeria clients are inadequate and that project teams allow clients to leave their briefing decision unresolved until the very late stages of projects. Clear and thorough project brief helps in clarifying the project objective to all project participants. This minimizes design error and non-compliance with the owner's requirements. Better initial planning came second among the three top-most important change order management strategies in building projects in northern Nigeria. This finding implies that project planning is not properly carried out in the Nigeria construction industry, and this could be a pointer to the fact that most errors found in contract documents result from inadequate planning. The finding also supports findings that project implementation is not correctly executed in most building projects in Nigeria.
Thorough detailing of drawing design was ranked the third most important change order management strategy in building projects in northern Nigeria where most contracts are based on drawings and specification. A well-detailed design is readily understood by construction professionals. This assists in identifying the errors, omissions and ambiguities in design at an early stage and helps avoid claims and disputes. This finding suggests that clients should allow sufficient time for designers to design, and also develop contract documents. This finding supports other researches, that building project in Nigeria are often hurriedly started without proper planning and as a result design consultants are put under pressure to deliver design drawings in order to meet up with deliver dates and in most led to lapses in design detailing. A possible explanation for this finding is that changes during design phase do not require any rework or demolition because no site work is involved. The multiple regression model has established the relationship between outcome and change order management strategies. The model in this study has provided empirical support that management strategies at design stage contribute most to the control of change order in building projects in northern Nigeria.

Conclusion
This paper was conducted to identify management strategies for building project and their relative importance and (ii) to determine the relationship between change order management strategies and building project success, by examining the perception of 323 respondents with different work experience and background in the Nigerian construction industry.
The findings show consensus of opinion among the respondents on change order management strategies.
Furthermore, the study found that change order management is 25.2% correlated with project success, but also shows quite clearly that 74.8% of project success cannot be explained by change order management. Based on the finding of this study the following recommendation are envisaged to help minimize change orders in project www.ccsenet.org/ass Asian Social Science Vol. 12, No. 1;2016 delivery in Northern Nigeria (i) Project sponsor need to properly adopt and use project implementation phase principles (ii) Increased effort at the design stages of projects.
(iii) Allocation of adequate resources to project requirement (iv) Adhere to budgeting provisions