Creating Effective Project Teams Using Personality Models
creating effective project teams using personality models Ki Young Jeong, Ph. D. , MBA. , University of Houston-Clear Lake Ipek Bozkurt, Ph. D. P. E. , University of Houston-Clear Lake Surya T. Sunkara, University of Houston-Clear Lake Hassan, Haider A. , Fairway Medical Technologies ____________________________________________________________ ________________________________ Abstract Human resources and team formation are important issues in any project success.
There's a specialist from your university waiting to help you with that essay topic for only $13.90/page Tell us what you need to have done now!
However, very little research was conducted in project team formation with consideration of diverse characteristics of human resources. Generally, human resources were assigned to project teams based on their availability and technical skills by a project leader. In this research, we considered personalities of human resources as an important factor in creating teams to maximize the team performance and/or to minimize the conflict within teams.
For this, three personality models – Myers-Briggs Type Indicator (MBTI), Fundamental Interpersonal Relations Orientation-Behavior (FIRO-B) and Kolbe Conative Index (KCI) – were discussed in the context of project team formation, and we concluded that both FIRO-B and KCI have advantages over MBTI in terms of project team formation since both models provide quantitative performance metrics while the MBTI just indicates types of personality. Finally, we presented a mathematical model to form effective teams using the results from the FIRO-B with a numerical example. Introduction
Forming effective project teams has been an interesting topic for many researchers and practitioners since human resources were considered one of the most important factors in affecting the performance of project teams. Traditionally, human resources have been assigned to project teams based on their availability and technical skills, and the interaction and personality aspects of team members have been often ignored or managed by project manager’s experience during this team assignment process. There is an enormous amount of research on human’s personality and its impact on team performance.
For example, Peeters et al. (2006) investigated the relationship between personality types in the ‘Big-Five’ model and team performance through meta analysis. According to them, both agreeableness and conscientiousness have a positive correlation with the team performance. Furnham and Stringfield (1993) used the Myers-Briggs Type Indicator (MBTI) to compare the managerial performance in Chinese and European cultures, and they found out that the MBTI generated significant differences between two cultures, however unrelated to robust and multi-factorial measures of management performance.
O’Neill and Kline (2008) used the ‘Big-Five’ to analyze the team performance results obtained from a business simulation game, and found out that both emotional stability and an individual’s predisposition to working in a team setting have a positive correlation with the team performance. LaFollette and Belohlav (1981) found out that there was no significant correlation between the team compatibility score computed by the Fundamental Interpersonal Relations Orientation-Behavior (FIRO-B) personality model and the simulated team performance.
However, many previous papers did not provide any systematic method to create effective teams based on the results from personality models. Recently, very few researchers started to investigate a systematic way to use personality models to create effective teams. Fitzpatrick and Askin (2005) provided a mixed integer programming (MIP) formulation to create effectives teams based on results from the Kolbe Conative Index (KCI) model. Andre et al. 2011) presented a formal model to assign human resources to teams where they use the Delphi technique to propose software project roles and competences, and used psychological tests including the MBTI and data mining tools to form a project team. The objective of this paper is aligned with that of Fitzpatrick and Askin (2005) and Andre et al. (2011). We discuss three personality models, MBTI, KCI, and FIRO-B to investigate their fitness to systematically create effective teams based on their results.
Specifically, we attempt to develop a mathematical model to form effective teams when the results from FIRO-B are given. The remaining of this paper is organized as follows: Personality models to discuss the MBTI, KCI and FIRO-B in the context of team formation, mathematical formulation to develop effective teams based on the results from personality models, numerical examples to demonstrate the mathematical models, and conclusions and future studies. Personality Models In this section, three personality models are briefly explained in the context of the project team formation.
MBTI. The MBTI is a personality model developed by Katherine Briggs and Isabel Myers based on Jung’s theory of psychological types, and it is the most popular personality type model in US (Gardner and Matinko, 1996). The MBTI uses the following four dimensions of preferences and each dimension has two opposite preferences: 1) Orientation of Energy: Extrovert (E) and Introvert (I). Extrovert people obtain energy from outer world while working with teams whereas introvert people obtain it from inside themselves. They may drain energy from interaction in a team. ) Manner of Information Gathering: Sensing (S) and Intuitive (N). People who prefer sensing rely on facts and reality, focusing on details while intuitive people rely on intuition, possibilities, and imagination, focusing on the big picture. 3) Manner of Decision-Making: Thinking (T) and Feeling (F). Thinkers are logical, analytical, and more objective than feelers. They make decisions based on principles, policies, and criteria. Feelers use values, emotion, devotion, and opinions from others. They want to maintain harmony within a team. 4) Attitude toward Structure: Judging (J) and Perceiving (P).
People who prefer judgment are outcome-oriented and decisive, and like closure and task completion. They tend to establish due date and stick to it while people who prefer perceiving are more process-oriented, and prefer to keep things open and flexible. They seek additional information and new possibilities. Based on two opposite preferences with four dimension, the MBTI generates total sixteen different preference type combinations (e. g. E or I; S or N; T or F; J or P). Hence a person may be identified to one of those sixteen combinations (e. g. ESTJ, ESTP etc).
Myers (1980) claimed that the best team performance is achieved by team members who differ on one or two preferences to complement each other, and who have two or three common preferences for better communication and understanding. She also recommended that types S and N are useful as well as types T and F. However, she claimed that people who have significant difference on J and P preferences will have difficulty to understand each other. Although the MBTI is a very useful instrument to form teams, it does not provide any specific quantitative metrics to be used in the team formation process.
Therefore, when project managers use the MBTI to select team members, they need to follow Myers’s recommendations mentioned above. In other words, it is still difficult to establish a systematic mathematical model to create teams with the MBTI since these recommendations do not completely characterize team formation processes. KCI. Although the KCI is considered a personality model in this study, it is slightly different from MBTI and FIRO-B in that it measures conative or individual’s instinctive behaviors that appear while that individual tries to achieve goals.
The KCI classifies instinctive behaviors into four categories: probing, patterning, innovating and demonstrating. Each category has its corresponding ‘Action Mode’ described below: 1) Fact Finder (FF): This is an action mode corresponding to the probing instinctive behavior. This mode is related to the information gathering. Fact finders are concerned with details, strategies, and research. They collect and analyze data, and establish priorities before making decisions. 2) Follow Through (FT): This is an action mode corresponding to the patterning instinctive behavior.
This mode is related to structure. People controlled by FT seek structure, and plan and schedules in advance. They also behave in a sequential manner. 3) Quick Start (QS): This is an action mode corresponding to the innovating instinctive behavior. This mode is related to the way to deal with risk and uncertainty. Quick starters initiate changes, take risk, and they are innovative. 4) Implementor (IMP): This is an action mode corresponding to the demonstrating instinctive behavior. This action mode is related to the way to handle space and tangibles.
Implementors are good at using space and materials, develop constructs, and easily use hand-on equipment. An individual has all of these four instincts to some degree, and can operate in any of these modes. Each of action modes can be classified into the following three operating zones based on the ten-point scale calculated from a set of questions: Prevent, Respond, and Initiate. 1) Prevent: A prevention-operating zone for an action mode has a score of 1 to 3. This zone basically represents the resistance or how the individual won’t act. ) Respond: A response or accommodating zone for an action mode has a score of 4 to 6. This zone represents how an individual is willing to act. 3) Initiate: An initiating zone for an action mode has a score of 7 to 10. This zone represents how an individual will act. In other words, the individual will tend to initiate that type of behavior and is comfortable working in that way. Let us suppose that each individual has its own KCI score. Then how to create effective teams? According to the KCI, team’s success or failure is dependent on the balance of conative energy inside the team.
A raw team score is developed by computing the distribution of team members across operating zones for each action mode. Then, the team synergy is calculated by computing the distribution of average percent of each zone across action modes. Exhibit 1 displays the team synergy values from an example in Fitzpatrick and Askin (2005). The ideal synergy in the final column is given by Kolbe Corp. through experimental evidence. The deviation from the ideal synergy – either positive or negative – is converted into the term profitability, a score out of one hundred.
A team with lower deviation from ideal team synergy values is considered to have higher synergy, resulting in high profitability. In general, a team with higher synergy is considered more successful since the team as a whole entity can use an appropriate amount of energy to initiate solutions, triggering appropriate actions required, and preventing further problems from occurring. For example, there is only one percent positive deviation from the ideal synergy in Initiate zone in Exhibit 1 while there is no deviation in Prevent and Response zones. Exhibit 1. Kolbe Team Synergy Zone/Action |FF |FT |QS |IMP |Team |Ideal | |Mode | | | | |Synergy |Synergy | | | | | | |(zone | | | | | | | |average) | | |Prevent |1 |7% |11% |62% |33% |28% |20-30% | |(30%) | | | | | | | | | |2 | | | | | | | | |3 | | | | | | | |Response |4 |14% |53% |20% |67% |41% |40-60% | |(60%) | | | | | | | | | |5 | | | | | | | | |6 | | | | | | | |Initiate |7 |69% |36% |18% |0% |31% |20-30% | |(30%) | | | | | | | | | |8 | | | | | | | | |9 | | | | | | | | |10 | | | | | | | When there is excessive conative energy in a combination of action modes and operating zones, the inertia occurs. According to Kolbe Corp, the ideal conative energy for both Prevent and Implement zones is 30% while it is 60% in Response zone as displayed in the first column of Exhibit 1. The positive deviations (excess values) of 12 action mode and zone combinations from the ideal conative energy value for zone are added to compute the inertia.
For example, the positive excess value (inertia) in Exhibit 1 is 86, given by [(62-30) + (33-30) for Prevent zone] + [(67-60) for Response zone] + [(69-30) + (36-30) for Initiate zone]. The KCI uses a term called, goal attainment, to measure the level of inertia in a team. The lower goal attainment means a stagnation of energy in a team. Therefore, it is apparent that we have to consider both team synergy and inertia when creating teams based on KCI. FIRO-B. The FIRO-B personality model was originally developed by Schutz (1966). It intends to measure the intensity of the interpersonal needs of inclusion (I), control (C), and affection (A). Inclusion refers to the need of individual’s social orientation – belongness and interaction.
Control refers to the need for power and influence, related to leadership behavior. Affection indicates the need for intimacy and friendship. Each of these needs has two dimensions to describe how much each of the three needs is expressed (e) or wanted (w). The expressed refers to the degree to which one behaves in that way toward others while the wanted describes the degree to which one wants others to behave that way toward oneself. The FIRO-B model describes the interaction of the three interpersonal needs with the expressed and wanted dimensions of each need, resulting in six categories of interpersonal needs as described in Exhibit 2. Each category is measured with a 10 point scale (0-9). Exhibit 2. FIRO-B Model Dimension|I |C |A | |/Need | | | | |e |The extent of |The extent of your |The extent of your | | |your effort to |effort to control |effort to get close | | |include others |and influence |to people, and to | | |in your |others. |engage them on a | | |activities. | |personal level. | |w |The extent of |The extent of your |The extent of your | | |your wish others|comfortability |wish others to act | | |to include you |working in well |warmly toward you. | | |in their |defined situations | | | |activities. |with clear | | | | |instructions. | | Suppose that we have FIRO-B scores for two persons, i and j.
Then, the following three different types of interpersonal incompatibility can be defined between these two individuals: Reciprocal incompatibility (RIij), Originator incompatibility (OIij), and Interchange incompatibility (IIij). Reciprocal incompatibility refers to the match between one’s need for expressed behavior and the other’s need for wanted behavior. This incompatibility occurs when one has a high level of expressed need while the other has a low level of the wanted need – see Equation (1) below, and any score higher than 6 indicates a strong possibility of incompatibility. Originator incompatibility indicates the match between two people’s expressed needs.
This incompatibility occurs when both want to initiate something or when neither wants to do – see Equation (2) below, and any score higher than 6 or lower than -6 indicates a strong possibility of incompatibility. Interchange incompatibility refers to the extent to which both have a similar total level of need in an area. For example, this incompatibility occurs when one emphasizes the affection need highly while the other emphasizes the control need highly – see Equation (3) below, and any score higher than 6 indicates a strong possibility of incompatibility. The formulae representing these three incompatibilities between two persons i and j are given in Equations (1), (2), and (3). RIij = |ei – wj| + |ej – wi|(1) OIij = (ei – wi) + (ej – wj)(2) IIij = |(ei + wi) – (ej + wj)|(3)
Exhibit 3 summarizes nine possible incompatibility combinations with their desired values. As seen in Exhibit 3, the row sum provides the overall reciprocal incompatibility (RI), overall originator incompatibility (OI), and overall interchange incompatibility (II) while the column sum provides the overall inclusion incompatibility (IIC), overall control incompatibility (CIC), and overall affection incompatibility (AIC). Finally, it also provides the total incompatibility (IC) as sum of all column sums or row sums. Please note that the desired score (threshold value) is displayed in the parenthesis. Exhibit 3. Summary of Incompatibility and Areas. Incompatibility |Areas |Row Sum | |Type | | | | |I |C |A | | |Reciprocal |RIij (+6) |RIij (+6) |RIij (+6) |Overall RI | |Originator |OIij (±6) |OIij(±6) |OIij(±6) |Overall OI | |Interchange |IIij(+6) |IIij(+6) |IIij(+6) |Overall II | |Column Sum |Overall IIC |Overall CIC |Overall AIC |Total IC | Therefore, the summary in Exhibit 3 suggests a way to create effective teams – create teams such that the resulting total incompatibility is minimized, which will be discussed in the next section. Mathematical Formulation
Based on the reason previously discussed, the MBTI is excluded from further consideration to form effective teams using quantitative approaches. Hence we discuss mathematical models using KCI and FIRO-B here. Suppose that we need to create multiple teams from a pool of workers with specific skills. Each team requires the specific number of employees with specific skills from the pool. Team formulation with KCI. Fitzpatrick and Askin (2005) developed the MIP model for the situation described above. i = a subscript for worker; j = a subscript for skill group; k = a subscript for team; z = a subscript for an operating zone – 1 for Prevent; 2 for Response; 3 for Implement; m = a subscript for an action mode – 1for FF;…,; 4 for IMP; [pic] Kj = set of workers in skill group j; [pic]
Sjk = number of workers of skill group j required for team k; Zz = desired level of zone z across all modes m (Z1 = Z3 = 1, and Z2 = 2); Zzm = maximum desired level of mode m at zone z (Z1m = Z3m = 30%, and Z2m = 60%); [pic]= negative deviation (deficiency) of team k’s synergy from an ideal value for operating zone z. For example, if team k’s synergy at Prevent zone is 15%, then [pic]= 5, negative deviation from 20. [pic]= positive deviation (excess) of team k’s synergy from an ideal value for operating zone z. For example in Exhibit 1, [pic]= 1 for team k since its Initiate synergy value is 31% while its ideal range is 20~30%; [pic] = positive deviation (excess) of team k’s energy from an ideal value for operation zone z and action mode m combinations.
The labor pool extraction model with multiple teams (LPEMT) is given by [pic](4) subject to: [pic] (5) [pic] (6) [pic] (7) [pic](8) for all variables ? 0; Please note that [pic] is a decision variable to decide which worker is assigned to which team. The first term in objective function (4) computes the deviations of team energy for all zones from ideal KCI values across all teams while the second term calculates the deviation (excess) of energy for all zone-mode combinations across all teams. Hence the objective function computes all deviations from optimal KCI values across teams.
Equation (5) ensures that each team has the appropriate number of workers from each skill group. Equation (6) ensures that a worker is assigned to at most one team. Equation (7) defines the deviation for each team from the optimal synergy by adding energy deficiency and by subtracting energy excess to and from the optimal synergy, respectively. Equation (8) defines the deviation for each team from the optimal inertia by adding energy deficiency and by subtracting energy excess to and from the optimal inertia, respectively. Team Formulation with FIRO-B. Now we consider the FIRO-B based LPEMT model to minimize the total incompatibility (LPEMT-TIC).
The following notations are additionally considered. i, j = subscripts for worker; l = a subscript for skill; r = a subscript for need (Inclusion, r = 1; Control, r = 2; Affection, r = 3); Slk = number of workers of skill group l required for team k; Kl = set of workers in skill group l; [pic] [pic]; [pic] [pic]; [pic]= reciprocal incompatibility of team k for need r; [pic]= originator incompatibility of team k for need r; [pic]= interchange incompatibility of team k for need r; Then the LPEMT-TIC is given, Min [pic]RIrk+ |OIrk|+IIrk )(9) Subject to: [pic] Yik[pic] 1, [pic](10) [pic]Yik = Slk, [pic](11) [pic] [pic](12) [pic] [pic](13) [pic] [pic](14) or all variables ? 0; The objective function (9) minimizes the total sum of three incompatibilities across all teams. Please note that we consider the absolute value of the interchange incompatibility in the objective function, and Yik is a decision variable. Again, equation (10) ensures that each team has the appropriate number of workers from each skill group, and equation (11) ensures that a worker is assigned to at most one team. Equations (12), (13), and (14) compute the RI, OI, and II across all teams for each need, respectively. That is, LPEMT-TIC model assigns each worker to each team such that the total incompatibility across teams is minimized.
Please note that LPEMT-TIC has nonlinear constraints and terms, [pic] for Equations 12 to 14. To avoid any technical complexity involved in solving these nonlinear constraints, we use Excel spreadsheet model with the Excel Premium Solver (EPS). Numerical Examples Two numerical examples are provided: One for LPEMT, the other for LPEMT-TIC. All data are randomly generated. KCI Team Formulation Example. The problem descriptions and results are displayed in Exhibit 4. We consider a pool of eight workers with two different skill groups, and want to create two teams where each team needs two workers from each skill group – Hence four workers each team.
The KCI profiles for two different skill groups (skill group 1 – workers 1, 3, 5, and 7; skill group 2 – workers 2, 4, 6, and 8) are provided for each operating zone (P – Prevent; R – Response; I – initiate) for each action mode. For example, a worker 1 has the (1, 0, 0; 0, 0, 1; 1, 0, 0; 0, 0, 1) profile, which may be rewritten to the zone-mode matrix below: | |FF |FT |QS |IMP | |P |1 |0 |1 |0 | |R |0 |0 |0 |0 | |I |0 |1 |0 |1 | Please note that there are 144 (= 3 x 4 x 3 x 4) different combinations of KCI scores equivalent to the values of this matrix – e. g. 827, 2718 etc where each digit represents KCI score of FF, FT, QS and IMP mode. For example, ‘3827’ is called a modus operandi (MO) score, representing 3 (P) in FF; 8 (I) in FT; 2 (P) in QS; 7 (I) in IMP. The shaded cells ‘N56’ to ‘O59’ (‘N56:O59’) and ‘N64’ to ‘O67’ (‘N64:O67’) display the output valued of decision variables obtained by EPS for skill group 1 and 2, respectively. Exhibit 4. LEPMT Excel Model Screen with Results from Premium Solver [pic] Please note that ‘1’ indicates that a corresponding worker is assigned to team 1 or 2. The cells ‘G75’ and ‘N75’compute deviations for zone-mode combinations for teams 1 and 2, respectively where its team synergy (TS) is also calculated for each zone across modes.
The cell ‘L79’ is the target objective function to be minimized by EPS, which is sum of cells ‘G75’ and ‘N75’. According to the results in Exhibit 4, team 1 consists of workers 1, 5 (from skill group 1) and workers 2 and 6 (from skill group 2) while the team 2 has workers 3, and 7 (from skill group 1) and workers 4 and 8 (from skill group 2), and this assignment generates 217 as its total deviation. FIRO Team Formulation Example. LPEMT-TIC based example for the same situation is displayed in Exhibit 5. The FIRO profiles for all eight workers are described in the cells ‘B5:G8’ and cells ‘B17:G20’ for teams 1 and 2, respectively. Decision variables are calculated in cells ‘H5:I8’ and cells ‘H17:I20’ for teams 1 and 2, respectively.
Based on the values in the decision variables indicating workers’ assignment to each team, three incompatibilities for two teams are calculated and displayed in cells ‘C27:F29’ for team 1 and cells ‘I27:L29’ for team 2. Their corresponding total team incompatibility is computed at ‘F30’ and ‘L30’ for teams 1 and 2, respectively. The minimized total incompatibility across teams is calculated at E35’. Exhibit 6 describes the EPS formulation using the spreadsheet model in Exhibit 5. For example, ‘E35’is set as an objective function to be minimized. ‘By changing variable Cells:’ pane displays the cells for decision variables. As explained in Exhibit 5, we have two decision variable cells: ‘H5:I8’ for skill group 1 and ‘H17:I20’ for skill group 2.
The first constraint in ‘Subject to the Constraints:’ pane is ‘H10:I10’ = ‘H11:I11’ to ensure that the number of workers assigned to each team, ‘H10:I10’, satisfies the required number of skill group 1 workers defined at ‘H11:I11’, representing that both teams need two workers from the skill group 1. The third constraint, ‘H22:I22’ = ‘H23:I23’, describes the same information for skill group 2. Exhibit 5. LPEMT-TIC [pic] Exhibit 6. Solver Formulation for LPEMT-TIC [pic] The second and fourth constraints, ‘H17:I20’ = binary, ‘H5:I8’ = binary, define that the decision variables are binary for skill groups 1 and 2, respectively. The fifth and seventh constraints, ‘J17:J20’