Dynamic Performance
Action Theory Analysis
(Frese & Zapf, 1994): Performance is not only determined by goals, action models of the task and the task environment, action plans, and feedback but high (and low) performance has an influence on these goals, action models, action plans, and feedback. When there is high performance, people achieve their goals in the action area A. As a result, they turn to other goals (often higher goals, Bandura, 1997) or to goals that are not in the same area. Thus, people may turn to an action area B. Once people turn to the other action area, they reduce attention to action area A; therefore, there is less learning in action area A and, the action plans may become less adaptive to the situation, because attention to feedback is reduced. This may lead to a performance plateau and may actually even reduce performance in action area A over time (Vancouver, 1997). The opposite effect appears when there is low performance, but high or medium aspirations: In this case, attention to achieving the goals in action area A is increased, including the development of better plans and better models of the environment and one's task structure, and better feedback processing. All these changes lead to higher performance which may be stopped when the goal of good (or adequate) performance is reached (and not higher aspirations appear).
Threat-Rigidity Cycle
(Staw, Sandelands, & Dutton, 1981) A threatening situation leads to more rigidity in the behavior; however, a higher degree of rigidity also leads to a higher performance, because rigidity reduces good information seeking and increases being reactive to the situation and not attempting to actively influence the situation. Such a reactive approach leads to low performance and is not a very effective strategy to deal with major problems at work (Van Gelderen et al., 2000). An example may be a study of small business owners: Lack of planning contributed to lower success levels (or higher failure rates) of these small businesses. Lack of success also led to a lower degree of planning (Van Gelderen, Frese, & Thurik, 2000). A positive cycle was also shown by Van Gelderen et al. (2000) that good action planning by business owners leads to higher firm performance, but that higher firm performance leads to better (and more) planning.
Affective Events Theory
(Weiss & Cropanzano, 1996) explains fluctuations in job performance by fluctuations in employee affect. More broadly, Affective Events Theory focuses on the structure, causes, and consequences of affective experiences and thereby explicitly acknowledges that affect fluctuates within the person. Work events are regarded as proximal causes of affective reactions that in turn influence behavior and attitudes. Thus, performance as behavior is seen as a consequence of affective experiences resulting from events encountered at work (along with mood cycles and dispositions to experience specific affective states). Weiss and Cropanzano (1996) summarized broad empirical evidence that affect predicts processes relevant for performance. More recent research that tested the theory's propositions about the correspondence between fluctuations in affect and fluctuations in performance is still rare, but generally supports the theoretical assumptions (Ilies et al., 2006; Rothbard & Wilk, 2006).
Suggestions for future Research
(a) advancing theory on dynamic performance (b) using a more systematic approach for studying performance trajectories over time (c) including contextual performance in the study of performance trajectories (d) paying more attention to performance fluctuation within persons.
Meta-Analysis Examining Performance Dynamics (Specific Results)
A breakdown of the results for specific subgroups of measures and time lags revealed that average stability of performance ranged between 0.44 (objective measure, low complexity, three-year time lag) and 0.96 (subjective measure, high complexity, 0.5-year time lag). Average temporal consistency ranged between 0.14 (objective measure, high complexity, three-year time lag) and 0.69 (subjective measure, low complexity, 0.5-year time lag). Average test-retest reliability ranged between 0.50 (objective measure, high complexity) and 0.83 (subjective measure, low complexity). Overall, these findings demonstrate that performance is far from being stable over time, but nevertheless seems to have a stable "component" as the stability coefficients decreased but did not approach zero. Interestingly, stability tended to be lower in highly complex jobs, a fact that can be attributed to a greater change in specific job requirements over time and an increased difficulty to assess performance when jobs are more complex.
Self-Efficacy Performance Cycle
A good example of reciprocal determinism is the effect of self-efficacy on performance and vice versa the effect of performance on self-efficacy (Lindsley, Brass, & Thomas, 1995; Shea & Howell, 2000). A variant of the self-efficacy-performance cycle is the high performance cycle described by Latham and Locke (2007). High and specific goals plus self-efficacy lead—via the mediators direction, effort, persistence, task specific strategies—to high performance, rewards, and satisfaction which, in turn, lead to higher commitment to the organization and to an increased willingness to accept challenges which in turn influence goal setting.
In What Ways Can Performance Fluctuate?
A person's performance levels may increase due to learning processes and practice and may decrease due to transient (e.g., fatigue) and more enduring (e.g., aging) processes. In addition, performance may also fluctuate within persons within short time intervals. If such fluctuations were random they might not justify much attention; however, if they are more systematic, it will be important to identify factors that predict a particularly high performance level within a person.
Short Term Changes Decreasing
A short-term decrease of knowledge, skills, ability, motivation, etc. can be observed as a result of fatigue. Mental fatigue can be defined as "a psychophysiological state resulting from sustained performance on cognitively demanding tasks and coinciding with changes in motivation, information processing, and mood." (van der Linden, Frese, & Sonnentag, 2003, p. 484). Fatigue leads to higher rigidity, lower degree of planning and a higher degree of errors as well as to a lower degree of motivation (Lorist et al., 2000; van der Linden, Frese, & Meijman, 2003). One of the results of fatigue is a reduction of attention (van der Linden & Eling, 2006). All of this leads to decreased performance. However, this decrease of performance is not absolute (and often not observable at work), because people often compensate the effects of fatigue with different strategies or enhanced effort (Meijman & Mulder, 1998; Sperandio, 1971).
Taxonomy- Long Term Changes / Skill Development
A similar process to the development of knowledge also appears in skill development; the increase of skills is an important part of expertise and there are interactions of skills and knowledge (Chase & Simon, 1973). An additional issue here is, however, that skill development if often developed in a tacit way (Myers & Davids, 1993) and, therefore, cannot be verbalized.
Taxonomy- Long Term Changes / Ability
Ability can increase and decrease depending upon the complexity of work—as discussed in this chapter. Age-dependent reductions of cognitive ability occurs more frequently in non-complex work (Schallberger, 1988; Schleicher, 1973; Schooler et al., 1999). Personality traits also change with age—often such a change seems to be related to maturity—older people tend to take less risks and are less open to experience. However, they are more likely to be conscientious and emotionally stable, "People become more socially dominant, conscientious, and emotionally stable mostly in young adulthood, but in several cases also in middle and old age. We found that individuals demonstrated gains in social vitality and openness to experience early in life and then decreases in these two trait domains in old age." (Roberts, Walton, & Viechtbauer, 2006). Also, wisdom increases with age (Baltes & Staudinger, 1993). Thus, maturation may be a natural process of time that affects both, ability and personality. Maturation also plays a role in emotional and motivational long-term changes.
Other Assertions Made About the Skill Acquisition Process
Ackerman (1988) further argued that the skill acquisition process is dependent on the complexity and consistency of the task. Complexity refers to memory load, number of response choices, amount of information provided to the learner and other aspects of the cognitive demands of the task. More complex tasks require more attention, reduce the accuracy of task performance, and increase the amount of time needed to complete a trial. Task consistency refers to invariant rules for information processing, invariant components of processing and invariant sequences of information processing components (Ackerman, 1987). Consistent tasks can become automatic, fast, effortless within rather short periods of time whereas inconsistent tasks can not be processed with automaticity and therefore, remain largely resource dependent. During the first phase, general mental ability is highly important, irrespective of task consistency. Consistent tasks become dependent on perceptual speed ability and psychomotor ability as practice increases; inconsistent tasks, however, remain largely dependent on general mental abilities because attention is needed for successful task completion.
Skill Acquisition Model
Ackerman 1987, 1988 This model differentiates between three stages of skill acquisition; each stage is associated with a certain type of abilities which predicts performance during this phase. The first phase of skill acquisition (cognitive phase) when the demands on the cognitive-attentional system are high, general mental ability is crucial for performance. During the second phase (associative phase) when the stimulus-response connections are refined, perceptual speed abilities are most relevant. Perceptual speed abilities refer to the "speed of consistent encoding and comparing symbols" (Ackerman, 1988, p. 290). During the third and final phase (autonomous phase) when tasks can be completed without full attention, psychomotor ability is most important for performance. Psychomotor ability describes the speed of responding to stimuli that involve no or only little cognitive processing demands. Thus, during the continuous process of skill acquisition the importance of general mental ability is high at the beginning and then decreases over time. Perceptual speed ability is low at the beginning, increases over time, and decreases again as task completion becomes more automatized. Finally, psychomotor ability is low at the beginning and increases over time.
Over and Under Confidence Limit the Spirals
Another mechanism that limits upward and downward spirals is caused by over- and underconfidence. Overconfidence may be the result of high performance (Lindsley et al., 1995) but it may lead to risky strategies that reduce performance again (Bandura & Locke, 2003; Vancouver & Kendall, 2006). Similarly, under-confidence is the result of low performance and may lead people to take extra care in preparing for an action (Sonnentag & Volmer, 2009), leading to higher performance. Of course, in any case, there must be a high aspiration level or strong incentives for high performance so that there is striving for high performance.
Changes in Cognitive Ability and the relationship of that change to performance
As a matter of fact, meta-analyses show that there is a zero relationship between age and core task performance (Ng & Feldman, 2008). A more differentiated analysis shows even some positive effects of age on reduced absenteeism, reduced tardiness, some forms of organizational citizenship behavior and small positive effects on safety behavior; counterproductive behavior and work place aggression are reduced with age (Ng & Feldman, 2008). The fact that age has no substantial effect on core job performance - in spite of decreased fluid intelligence as a function of age—can be explained by the complex nature of the performance dynamics.
Research linking individual performance to team-level processes
Chen (2005) explicitly linked individual performance trajectories to team-level processes. More specifically, he analyzed performance change of knowledge workers after they joined a project team. Performance was assessed via peer and team leader ratings. Analysis showed that initial team performance predicted change in newcomers' performance over a period of six weeks. Although empowerment and team expectations predicted newcomers' initial performance levels, there were no significant predictors for performance change. This study is noteworthy as it included situational variables (e.g., team performance) as potential predictors of performance change. In combination with the other studies, it suggests that not only individual-difference variables, but also environmental variables play a major role for explaining differences in performance trajectories.
Changes in Cognitive Ability over the Lifespan
Cognitive ability and conscientiousness change across the life span. Particularly the fluid intelligence part of cognitive ability is reduced over time, while crystallized intelligence does not decrease (Baltes, Staudinger, & Lindenberger, 1999; Verhaeghen & Salthouse, 1997). The biological process of a sharp reduction of all parts of intelligence comes at a very high age (when most people are retired from work; Baltes et al., 1999). However, there is a certain degree of plasticity even in the decrement of intelligence over the life span (Baltes et al., 1999). Work psychology also produced evidence that the reduction of cognitive ability is dependent upon the complexity of work done; employees in highly complex jobs show few signs of reduction of cognitive ability with age in contrast to employees in less complex jobs (Kohn & Schooler, 1978; Schallberger, 1988; Schleicher, 1973; Schooler, Mulatu, & Oates, 1999). This effect of complexity of work is more pronounced in older than in younger workers (Schooler et al., 1999).
Individual Differences Related to differences in Performance Trajectories
Deadrick et al. (1997) examined performance trajectories of sewing machine operators over a relatively short period of 24 weeks. job experience was negatively related to the linear growth parameter indicating that workers with previous experience in the specific domain improved more slowly than workers with no previous experience. Cognitive ability predicted a fast increase in performance. Interestingly, only 5% of the variance in the linear performance trend was explained by the predictors included in this analysis. Ployhart and Hakel (1998) analyzed gross sales commissions of salespeople over a period of eight quarters. Latent growth curve modeling demonstrated that past salary and future expected earning predicted interindividual performance differences at the first measurement occasion. Self-reported persuasion and self-reported empathy were related to the linear growth parameter, indicating that those who think of themselves as persuasive and empathetic increase their sales performance at a faster pace than those who rate themselves lower in these dimensions. The predictors were not related to a quadratic growth parameter (i.e., acceleration in performance over time).
Examining Within Person Fluctuations in Engagement in Action Phases
Examining within-person fluctuations in engagement in action phases (Frese & Zapf, 1994). Fluctuations in task requirements and in affect might be responsible for such fluctuations. For example, goals may change when they are achieved, plans are adjusted when barriers, problems or errors appear, and feedback processing changes with environmental conditions. Also, when people are in a negative affective state, they tend to process information more systematically than in a more positive affective state (Schwarz & Bohner, 1996). Therefore, one can assume that planning or feedback processing will increase with negative affective states.
Why is there no relationship with changes in age and job performance?
First, some of the jobs may require little fluid intelligence even though crystalline intelligence requirements may be high (e.g., managers; Kanfer & Ackerman, 2004). Second, there are compensatory mechanisms. People at work may compensate memory loss by writing things down, optimize their approaches to the tasks (effort, time allocation, etc.), and select (if possible) those tasks that they are good at (or reduce the number of tasks)—all this suggests a SOC (selection - optimization - compensation) approach to performance is functional, particularly at old age Third, workers may be able to compensate memory losses with increased levels of conscientiousness. There is evidence that older workers become more conscientious across the life span (Roberts, Walton, & Viechtbauer, 2006). Conscientious older workers use SOC approaches (Baltes, 1997) to keep up good performance (Bajor & Baltes, 2003). Thus, good predictors of performance (such as conscientiousness) tend to change over time. It is useful to think of motivational effects of these various changes as mediating mechanisms of performance change over time (Kanfer & Ackerman, 2004).
Two Issues with the Deviation Amplification Hypothesis
Firstly, as Shea and Howell (2000, p. 791) point out, that the self-efficacy-performance cycle is probably not a never ending positive cycle upwards, because "the pattern of changes in self-efficacy and performance from trial-to-trial contained self-corrections, suggesting that the efficacy-performance relationship does not necessarily proceed in a monotonic, deviation-amplifying spiral." Thus, there are self-correcting, self-regulating processes that lead to asymptotes rather than to never ending positive or negative cycles. Secondly, as of yet, there are no good data that suggest a variance increase across time. For example, the study on the cycle of control and complexity at work and personal initiative did not show any increase in variance of performance (and in the predictors) over time (Frese et al., 2007)
Predictors of Short Term Performance Fluctuations
Fisher and Noble (2004) found that momentary task performance was predicted by perceived skill level (with respect to the specific task), task difficulty, interest and effort. Effort partially mediated the effect of interest on task performance. Ilies et al. (2006) showed that momentarily experienced states (positive affect, job satisfaction) predicted day-specific organizational citizenship behavior. On days when employees experienced high levels of positive affect and job satisfaction, they engaged more in OCB. Ilies and his coworkers (2006) further tested interaction effects between the experienced states and personality variables. A test of cross-level interactions showed that high levels of agreeableness attenuated the effects of positive affect on OCB. Persons with high agreeableness showed relatively high levels of OCB irrespective of their momentary affect, whereas persons with low agreeableness showed OCB when experiencing high levels of positive affect, but not when experiencing low levels of positive affect
Long Term Changes Decreasing
Forgetting is an important process that leads to a reduction of knowledge and skills. Forgetting may be a function of time - decay (Portrat, Barrouillet, & Camos, 2008) or of interference of new information (Oberauer & Kliegl, 2006). For skills, nonuse may be most important for its loss. Finally, burnout effects in the sense of exhaustion (in contrast to energy), cynicism (in contrast to involvement), and inefficacy (in contrast to efficacy) may also be a function of time in a specific and stressful environment (Maslach & Leiter, 2008).
Curvillinear Effects: Habit Regression
Habit regression appears when two conflicting habits have been developed sequentially (one habit is older than the other) and some frustration or conflicting environmental cues are added. This often leads to regressing (reverting back) to the older habits (Mowrer, 1940). This is important at work when workers revert back to older habits under time pressure or when people are again in the environment in which old habits had developed. This phenomenon is similar to regression because the old habits were originally developed in the work environment; in this case, the new behavior may also not have been developed into a habit yet (i.e., these newly acquired behaviors have not been routinized yet). In this case, the conscious decision to use the new behaviors learned in the training program cannot be put into effect because of cognitive interference due to double tasks or time pressure. Thus, relapse to old habits ensues. Similar concepts, such as rigidity, resistance towards change in organizational development (Cummings & Worley, 1993), structural inertia of organizations (Aldrich, 1999), entropy in system theory (Katz & Kahn, 1978) have attempted to describe something similar—the tendencies of individuals and organizations not to change although the need to change may be accepted—probably, there is a large overlap of these terms with relapse to prior habits and routines.
Within Person Variability in Contextual Performance (Research)
Ilies et al. (2006) analyzed 825 experience-sampling data points collected from 62 persons over a period of 15 working days. They found that 29% of the total variance in OCB resided within persons. Binnewies et al. (2009) using day-level data from 99 persons reported that even 50% of the total variance in OCB was within-person variance. In the study by Binnewies et al. (2009), 56% of the total variance of personal initiative behavior resided within persons. Also proactive behavior such as personal initiative fluctuates within person. In a day-level study on personal initiative as a specific type of proactive behavior, Sonnentag (2003) found that 46% of the total variance was within-person variance.
Performance on Specific Days Matters
In many modern workplaces it is not enough to show high performance on average, but to perform reliably well on specific days (Sonnentag, Dormann, & Demerouti, 2010). Examples for the importance of having to perform reliably are the implementation of a new technological system within a very short time or a presentation to be delivered to a very important customer. If individual employees' performance is low on these days, negative consequences, both for the organization and the individual, may be severe—no matter how good employees' average performance is. Thus, it is crucial that on such days, individuals perform at a high level (i.e., better than their mean level).
Motivation and Habit Regression
In motivation, this relapse to older habits plays a large role as well (e.g., if one is not really motivated to show the new behavior at the work place). In addition, a motivational theory, called the opponent process theory of motivation might be used to explain curvilinear changes as inherent processes of motivation and emotion: "The theory assumes that many hedonic, affective, or emotional states are automatically opposed by central nervous system mechanisms which reduce the intensity of hedonic feelings, both pleasant and aversive. The opponent processes for most hedonic states are strengthened by use and are weakened by disuse." (Solomon & Corbit, 1974, p. 119). This theory can also explain how adaptation processes appear so that cycles do not continue forever.
Interpersonal Predictors of Performance Fluctuations
In their study of sales performance, Stewart and Nandkeolyar (2006) identified referrals (i.e., specific information about sales opportunities) from a central office as an important predictor of a sales person's week-level performance. More than 60% of the variance in a sales person's weekly performance was explained by referrals from the central office. Moreover, personality variables moderated the relationship between referrals and performance; the relationship was stronger for persons high on conscientiousness and lower for persons high on openness for experience. Following this line of research, Stewart and Nandkeolyar (2007) examined weekly performance of professional football players and found that also in this domain performance fluctuation can be explained by constraints from outside the focal performing person. More specifically, teammate constraints (i.e., teammates competing for individual performance) and opposition strength were negatively related to weekly performance.
The Dynamics of Performance are Different for Different People at Different Times
Instability may not be uniform for all individuals and individuals may differ in their changes over time.Several studies have examined predictors of interindividual differences in performance trajectories and whether performance trends are systematic Hofmann et al. (1993) found that individuals differed in their performance trajectories, that is the change in performance over time was not uniform across individuals. More specifically, 69% of the variance of linear growth parameter and 30% of the variance of the quadratic growth parameter were systematic, indicating that linear and quadratic parameters differed between individuals; however, there were no differences between individuals for a cubic parameter. Hofmann et al. (1993) suggested that differences in individual skills, knowledge or goal orientations might influence differences in performance trajectories. Research has found that performance can be linear, quadratic and cubic. Quadratic trends imply that the increase in performance decreases over time (i.e., "flattens out"); cubic trends refer to more complex patterns that often can only be interpreted in the light of the specific research setting and the specific time frame (cf., Thoreson et al., 2004). Rarely studied.
Equilibrium in the cycles or spiral
It is also possible that dynamic processes lead to some kind of equilibrium or some pendulum movement. For example, if high performance has negative effects on the predictors of performance and if low performance has positive effects on these predictors, some equilibrium or possibly a pendulum movement will appear (Vancouver, 1997). An example would be that job satisfaction leads to higher performance, but performance reduces job satisfaction because too much motivated effort is necessary to keep up high performance, which may have detrimental effects on job satisfaction.
Other Ending Considerations
It would be interesting to examine how engagement in various performance phases (e.g., planning, feedback processing) unfolds its dynamics. For example, explicit planning probably decreases over time. Moreover, changes in the performance process might be related differently to performance outcomes at various stages of a performance trajectory. Most studies on performance trajectories did not provide much information about HR interventions that might have occurred during the study time. For example, trainings and other HR practices might boost performance levels over time. Sturman (2007) suggested that future studies might want to explicitly look at the effects of HR interventions on performance trajectories Should examine also counterproductive behavior from a dynamic perspective Very little is known about the boundary conditions of performance variability within persons. For example, in highly structured work settings with low levels of job control performance variability might be much smaller than in settings where employees can decide themselves about how to do their jobs (cf., Binnewies et al., 2009).
Deviation Amplification Hypothesis
Lindsley et al. (1995) CONTROVERSIAL A deviation-amplifying loop exists because "a deviation in one variable (decrease in self-efficacy) leads to a similar deviation in another variable (lower performance), which, in turn, continues to amplify. Thus, the cyclic nature of the self-efficacy-performance relationship can result in a downward (decreasing self-efficacy and performance) or upward (increasing self-efficacy and performance) spiral."
Taxonomy- Long Term Changes / Expertise
Long term changes that increase or lead to a stronger expression of a personality trait or an emotion may appear in knowledge, skills, ability and personality, emotion, and motivation. Knowledge increases as a result of learning processes. Much like expertise is developed over time, so is knowledge developed as a result of doing something for a longer period of time (Tesluk & Jacobs, 1998). However, research has persuaded us that expertise is not purely a function of time (in the sense that people increase their knowledge because they are doing something longer), but rather, expertise develops as a function of the time people spend in processing knowledge in depth. For example, it is not true that purely spending time as a programmer produces expertise in programming; rather intensity, breadth, and depth of programming leads people to be called experts by their peers (Sonnentag, 1996, 1998). This type of expertise can be developed by deliberate practice, for example, with a help of a coach (Ericsson & Lehmann, 1996) or when people force themselves to work on issues that are difficult and that lead to a maximal increase of knowledge and expertise (Sonnentag, 2000; Sonnentag & Kleine, 2000; Unger, Keith, Hilling, Gielnik, & Frese, 2009)
What is Job Performance?
On a basic level, job performance comprises a process aspect (i.e., behavior) and an outcome aspect (Campbell, McCloy, Oppler, & Sager, 1993; Motowidlo, 2003; Roe, 1999). The process aspect refers to multiple, discrete behaviors that people do at work (Campbell, 1990). It focuses on the action itself—as opposed to the results or outcomes of this action. It limits itself to behaviors that contribute to the goals of the organization and that are under the control of the acting person (Campbell et al., 1993). The outcome aspect refers to the results of behavior and comprises states or conditions that are only partially under people's control. While the performance process aims to achieve positive performance outcomes, the performance outcomes (i.e., results) are usually influenced by other processes than only a person's performance behavior (e.g., situational and organizational constraints, market conditions, random processes).
Personal Initiative's reciprocal relationship with work characteristics
People are supposed to show personal initiative when they can influence conditions at work (control) and when they have the required competencies (resulting from complexity). Thus, control and complexity at work should increase personal initiative. In turn, personal initiative should also lead to increasing control and complexity, because people with high personal initiative may generate some added control and complexity in their given jobs. Tasks of job are not fixed because of emergent elements in a job (person could be good a job due to initiative, boss gives them more responsibilities thus the tasks may change). Job change may occur. People high in personal initiative are likely to look for and make use of opportunities for getting more challenging jobs and for increasing their career success. These predictions were borne out by the data in a longitudinal study with four measurement points (Frese et al., 2007)—this process is mediated by control aspiration (desire to exercise control), perceived opportunity for control (expectation to have control), and self-efficacy (belief in own competence).
Research on Performance Cycles in 3 Areas of IO
Performance cycles have been shown in three areas: Firstly, the high performance cycle of goal setting research has been shown to exist. Here, high and specific goals and high self-efficacy lead to high performance which, in turn, affects rewards and satisfaction, which, in turn, increases commitment which in turn affects the willingness to take on challenges of high and specific goals (Latham & Locke, 2007). Secondly, there is good data showing that self-efficacy leads to higher performance which, in turn, leads to higher self-efficacy (Lindsley et al., 1995, Frese et al., 2007). Finally, personal initiative and engagement as active performance strategies have been shown to affect the work environment so that higher challenges appear. This in turn then leads to higher personal initiative and engagement (Frese et al., 2007). The common theoretical concept of all performance cycles is that an active form of performance or high performance either changes the challenges posed by the work environment, or increases the readiness to accept challenges which in turn has an influence on these active approaches to performance.
What is Dynamic Performance?
Performance is considered to be dynamic if it changes over time without outside and directed interventions; changes over time can imply changes in mean values, changes in correlations between performance dimensions, and lack of stability of job performance over time (Deadrick, Bennett, & Russell, 1997; Hanges, Schneider, & Niles, 1990). According to Sturman (2007), lack of stability refers to the behavioral aspect of performance, not to the results or utility of performance. One implication of this specification is that individuals' rank order of performance scores change over time. Change may refer to intra-individual change and to interindividual differences in intra-individual change (Hofmann, Jacobs, & Baratta, 1993).
Personal Initiative's reciprocal relationship with work engagement
Personal initiative means that people are self-starting (goals are developed and pursued without external pressure, role requirements, or instruction), proactive (prepared for future negative or positive events), and are persistent in overcoming barriers and problems (Frese & Fay, 2001). One set of studies examined longitudinally the effects of job resources on work engagement (composed of vigor, dedication, absorption), of work engagement on personal initiative and of personal initiative on work engagement, and of work engagement on job resources (Hakanen, Perhoniemi, & Toppinen-Tanner, 2008). The overall results imply that job resources increase the level of work engagement and that there are reciprocal influences of work engagement and personal initiative.
Dynamic Performance may refer to short and long term changes
Reb and Cropanzano (2007) characterized the long-term changes as performance trends. Such long-term changes result—among others—from changes in knowledge, skills, and experience. In organizational settings these trends may cover periods of months or even years (Deadrick et al., 1997; Ployhart & Hakel, 1998), in laboratory research such trends occur over periods of some hours (Ackerman, 1988). From such performance trends, short-term performance fluctuations can be differentiated, reflecting performance variations around a constant mean (Reb & Cropanzano, 2007). These variations are not random but may be influenced by momentary affective states (Beal, Weiss, Barros, & MacDermid, 2005).
Reciprocal determinism
Reciprocal determinism implies that "people are both, producers and products of social systems" (Bandura, 1997, p. 6). People have an influence on the surrounding social systems (e.g., the work group, the organization, the supervisor, the division of labor and work place) which, in turn, has an influence on how they behave. Performance cycles should appear whenever performance changes those conditions that have an influence on performance. This is particularly so for psychological constructs, which are instrumental in changing conditions. Thus, active forms of performance are more likely to lead to such changes in conditions which, in turn, change the active form of performance.
Within Day Fluctuations
Research has shown that performance parameters fluctuate across the day (Daniel & Potašová, 1989; Folkard, 1990). One of the earliest studies of the circadian rhythm was on errors (often conceptualized as the converse of performance; Bjerner, Holm, & Swensson, 1955).
Within Person Variability in Task Performance (Research)
Research using experience sampling methodology (Beal & Weiss, 2003; Reis & Gable, 2000) and similar approaches (including daily and weekly surveys) provide rather consistent evidence that performance fluctuates substantially within persons. For example, Stewart and Nandkeolyar (2006) examined weekly sales performance of 167 sales persons over a period of 26 weeks and found that 73% of the variance in performance resided within persons. Similarly, a study on performance of professional football players revealed that 63% of the week-to-week variance in performance was within persons (Stewart & Nandkeolyar, 2007). Trougakos et al. (2008) found that even within small time units, performance varies largely within persons. Within-persons variance in observed performance during a total of eight performance episodes during three days was 48%. Fisher and Noble (2004) reported substantial fluctuation of performance within persons. Specifically, in an experience sampling study with five measurement occasions per day, these authors found that subjectively rated performance at a given point in time only predicted 3% of subjectively rated performance at the following measurement occasion.
SOC (selection - optimization - compensation) approach
SOC (selection - optimization - compensation) approach to performance (Baltes, 1997; Zacher & Frese, 2009). Selection (select those activities and goals that are central), Optimization (e.g., seizing the right moment, investing resources, honing skills, knowledge, and attention for those activities that were selected), Compensation (maintaining good performance in spite of loss by making good use of alternative means) are functional, particularly at old age
Cycles Taxonomy
Self-efficacy leads to higher performance; it also leads to an increase of the complexity and controllability of the work tasks which, in turn, lead to higher self-efficacy (Frese et al., 2007, have shown such job enrichment as a result of control orientation (which includes self-efficacy) among others). Upward and downward cycles also exist in the areas of ability and motivation. The "broaden and build" theory of positive emotions is related to such cycles; it shows positive emotions to have a positive influence on that type of performance that needs to incorporate new information (such as innovation tasks; Fredrickson & Losada, 2005). The emotional cycle view assumes that the "broaden and build" theory of emotion leads to more challenging work (because people are open to innovations); good performance in innovative tasks leads to positive affect which, in turn, would increase openness to innovation tasks (Fredrickson & Losada, 2005). In the field of motivation, there is both the self-efficacy-performance cycle as well as the personal-initiative—work characteristics cycle (control and complexity). Both of these have been described above. The downward cycles can all be the obverse of the above effects—thus, low self-efficacy can cycle downward because people do not take up challenges, do not work hard on complex and controllable tasks and, therefore, the challenges are reduced which leads to low self-efficacy. The opposite of personal initiative is a reactive strategy which leads to lack of success, which, in turn, leads to a higher degree of reactive strategy in small business (Van Gelderen et al., 2000). A specific negative cycle is the threat-rigidity cycle: individual or organizational threat leads to individual or organizational rigidity which produces reactive approaches, which, in turn, leads to higher threat (Staw et al., 1981).
Recovery-Related Predictors of Performance Fluctuations
Specifically, Sonnentag (2003) and Binnewies et al. (2009) Using a daily-survey design, the studies showed that feeling recovered in the morning predicted task performance (Binnewies et al., 2009), OCB (Binnewies et al., 2009), and personal initiative (Binnewies et al., 2009; Sonnentag, 2003) throughout the day. Mediation analysis identified day-specific work engagement (Schaufeli & Bakker, 2004) was the mediator between feeling recovered in the morning and personal initiative behavior. Probably, feeling recovered indicates that regulatory resources are available, which can be invested in the task accomplishment process. A recent study by Trougakos et al. (2008) supports this view. This study focused on one specific type of task performance, namely affective display in cheer leaders. Using an experience sampling design, Trougakos and his coauthors showed that positive emotions and respites (i.e., breaks) from previous work efforts were positively related to subsequent task performance.
Learning Curve Theory
Sturman (2007) This modeling approach states that when a task is repeatedly executed over time, task performance improves and that this improvement can be represented by specific mathematical models. Although Learning Curve Theory has been developed to predict organizational productivity, it might be useful for specifying also the shape of individual performance over time. Sturman (2007) discussed necessary steps to be undertaken before Learning Curve Theory can be fully applied to the prediction of individual performance changes over time. These steps include the development of the appropriate functional form for modeling performance over time, the extension to non-routine tasks and including the effects of management efforts to increase performance
Outcomes of Individual Differences in Performance Trajectories
Sturman and Trevor (2001) Hierarchical linear modeling revealed that persons who stayed in the organization during the eight-month study period showed a positive performance trend over time; however, persons who left the organization during the study period, had not increased their performance. When current performance level was controlled, longer-term negative performance trends in the past predicted turnover. Moreover, longer-term performance trends in the past interacted with current performance in predicting turnover: employees with low current performance levels showed an increased tendency to voluntary turnover when their performance trend in the past was negative, but not when their performance trend in the past was positive; for employees with high current performance, past performance trends did not matter with respect to voluntary turnover. The evidence on nondynamic models of turnover may point to conclusions that are only related to the current performance levels to be predictors of turnover (Williams & Livingstone, 1994) and not to past trajectories of performance. It would be an interesting avenue for future research to examine performance trends related to other variables such as job attitudes or to motivational constructs including self-efficacy. For example, research within a control theory framework suggests that not just a person's current performance level and progress towards a goal, but also the velocity of goal attainment predict affective outcomes (Lawrence, Carver, & Scheier, 2002). Similarly, longer-term performance trends that reflect these velocity aspects might be relevant for job satisfaction and similar constructs.
Meta-Analysis Examining Performance Dynamics
Sturman et al. (2005) addressed these issues of performance dynamics in a meta-analysis and examined the extent of job performance changes over time. This meta-analysis based on 22 independent samples with an overall sample size of 4,294 individuals made use of a total of 309 correlations with time lags ranging between one week (Rothe, 1970) and 72 months (Hanges et al., 1990). Specifically, they aimed at separating stability (extent to which the true score does not change over time) from temporal consistency (extent to which observed performance measures correlate over time) and test-retest reliability (degree of convergence between the observed and the true score). Overall, this meta-analysis showed that temporal consistency decreased as time between measurement points increased and that objective performance measures (as opposed to subjective measures) and greater job complexity were associated with lower test-retest reliability, with job complexity moderating the effects of time.
Types of Process Performance
Task performance refers to activities that directly transform materials or information into goods or services and to activities that directly support the organization's core activities (e.g., by delivery activities or planning and coordination; Motowidlo, 2003). Contextual performance refers to discretionary behaviors that add to organizational effectiveness by improving the functioning of the social and organizational context of work. Adaptive performance concept is that people cope with and support organizational change (Griffin, Neal, & Parker, 2007; Pulakos, Arad, Donovan, & Plamondon, 2000). The proactive performance concept (Grant & Ashford, 2008) refers to behaviors that initiate change, are self-starting and future oriented (Frese & Fay, 2001; Griffin et al., 2007). It implies personal initiative (Frese, Kring, Soose, & Zempel, 1996; Thompson, 2005), taking charge (McAllister, Kamdar, Morrison, & Turban, 2007; Morrison & Phelps, 1999), voice (LePine & Van Dyne, 2001), active feedback seeking (Ashford & Tsui, 1991) and some forms of engagement (Macey & Schneider, 2008).
The changing-subject model
The changing-subject model (changing-person model as described by Keil & Cortina, 2001) proposed that individual characteristics relevant for performance change over time as the individual gets more experience with the task. One specific formulation of this approach referred to changing levels of abilities (Adams, 1957). Later, it was argued that it might be more appropriate to specify changes in skills because skills are conceptualized to be more changeable than abilities (Keil & Cortina, 2001). Sturman (2007) argued that also motivation and job knowledge may change over time and could therefore be incorporated in changing-subject models. It is important to note, that this model assumes that the causes of performance change over time, but that the contribution of these causes to performance may remain stable
The changing-task model
The changing-task model states that the contribution of specific abilities to performance changes over time, but the individual's level of these abilities remains stable. The relative contribution of specific abilities may change over time because of job changes, assignment of new roles and tasks and because of revised organizational requirements such as changes in technology (Sturman, 2007). Keil and Cortina (2001) have argued that these two models should be seen as complementary rather than competing.
Auto Regressive Components
The difference between dynamics and logitudinal. Performance is a good example of a variable with an autoregressive (you ar eregressing the time t value on the time 1 value) time 1 predicts time 2, population grouth is an example of explosive growth Dynamics will have a steady state which is where a thing reaches asymotpes. The OCB Latent growth curve is just a trend can be linear, quadratic, it implies that it will go in a certain direction forever. However, in real life, it will probably reach equillibrum rahter than going up or down endlessly. On average you will be on a certain setpoint Variable can predict equalibrium (like conflict example) mediators can cause a new equillibrum Lag length- does the time one measurement effect measurement at time 2 or time 3? Often in the cross-sectional study Note that time is not a causal variable!!! But when you lok at the latern growth curve line, it includes time in the formula that is the difference between this and regular regression LATENT GROWTH CURVES ARE NOT DYNAMIC THEY ARE LONGITUDINAL THings that update selves- affective states, cohesion
Which Areas is Dynamic Performance Relevant to?
The dynamics of performance is particular evident in the following areas of performance: Learning and forgetting, temporal vigor and fatigue, engagement and burnout
Performance Fluctuation and the Circadian Rythym
The important issue is that performance fluctuations are regulated by the circadian rhythm; this means that humans typically show a higher performance at certain times across the day (e.g., between 6:00 and 9:00 in the morning or at around 6:00 in the afternoon). The lowest level of performance is likely to be at around 3:00 in the night. The interesting finding here is that these effects appear regardless whether people have already worked for eight hours or are just starting their work day; there seems to be a seamless performance cycle as a result of the circadian rhythm (Bjerner et al., 1955).
Predictors and Performance: Which one affects the other?
The literature often assumes that only the predictor of performance affects performance and that there is no further effect of performance on these predictors. However, theoretically, hypotheses can be developed in both directions If high performance has positive effects and low performance negative effects on these predictors, then positive (virtuous) or negative (vicious) cycles will appear. An example would be that positive emotions have an effect on performance and that performance leads to more positive emotions, e.g., job satisfaction (Judge, Thoresen, Bono, & Patton, 2001).
Personality Related to differences in Performance Trajectories (Maintenance Sample)
Thoresen and his coauthors built on the work of Murphy (1989) and distinguished between maintenance and transitional job stages. Personality seems to matter for changes in the transitional stage, but not for changes in the maintenance stage. For the maintenance sample, the authors found the following: There was a significant linear and cubic trend indicating an overall increase in performance over time with a large increase in performance from Quarter 1 to Quarter 2, relative stability from Quarter 2 to Quarter 3, and another increase from Quarter 3 to Quarter 4. Also, job tenure and conscientiousness were significantly related to mean levels of performance; extraversion was a marginally significant predictor of performance. Additionally, in this maintenance sample, there were no significant interactions between Big Five personality variables and the linear and quadratic trends in the performance trajectories, suggesting that broad personality constructs did not matter much with respect to performance increase and acceleration over time during the maintenance stage. However, there was a significant interaction between conscientiousness and the cubic trend of sales performance indicating that higher-order patterns of performance trajectories may depend on conscientiousness in the sample that had already developed a certain number of routines.
Personality Related to differences in Performance Trajectories (Transition Sample)
Thoresen et al., 1989 For the transitional sample, analyses revealed: There was a strong positive linear trend and a significant negative quadratic trend (but no cubic trend) indicating a strong general increase in performance and a plateauing of this increase over time. Also, agreeableness and openness to experience were significantly related to mean performance also when job tenure was controlled for. Moreover, agreeableness was positively and emotional stability was negatively related to the linear growth parameter. This unexpected finding with respect to emotional stability might indicate that employees low on emotional stability react more strongly to disturbances and unforeseen changes, which in turn might help to increase performance. Openness to experience was negatively related to the quadratic growth parameter indicating that persons high on openness were less likely to show plateauing of their performance level over the study period.
Why is it important to consider performance dynamics?
To arrive at a better understanding of job performance at the conceptual level, research attention must be paid to the dynamic aspects of performance. Researchers agree now that performance and performance criteria can change over time. Taking a dynamic approach to performance enables researchers to make predictions about performance trajectories over time. Taking a dynamic approach to performance, makes it possible to combine research on performance with research on life long development, thus, integrating knowledge about performance changes across the life time into work and organizational psychology (Kanfer & Ackerman, 2004; Schalk et al., in press) Acknowledging the dynamics of performance makes it possible to examine fluctuations of performance around a person-specific mean level (Reb & Cropanzano, 2007). We are then able to develop indicators about the degree of these performance fluctuations (e.g., within a week, a month or a year) and whether they are still in a "normal" range or not HUGE RAMIFICATIONS FOR SELECTION: as research on dynamic performance has shown, rank orders are not completely invariant over time (e.g., (Hofmann, Jacobs, & Gerras, 1992). Thus, the utility of such traditional selection approaches that assume an unchanged rank order of individuals' performance over time cannot remain unquestioned. To improve selection procedures, research may identify factors that predict inter-individual differences in performance trajectories (i.e., individual differences in performance change over time; cf., Ployhart & Hakel, 1998).
Episodic Performance and Affect's Effect on Performance
To explain within-person fluctuations of performance, Beal et al. (2005) proposed a theoretical model that focuses on episodic performance. Performance episodes are relatively short units of behavior that are naturally segmented and organized around rather proximal work-related goals. Typically, one workday comprises several performance episodes. By building on resource allocation models (Kanfer, Ackerman, Murtha, Dugdale, & Nelson, 1994), Beal et al. (2005) suggested that performance within each episode is influenced by the person's general resource level (e.g., cognitive ability, task-relevant skills) and the momentary allocation of resources. Performance within an episode is impaired when the person does not succeed in allocating all necessary resources to the primary work task and when attention is diverted by off-task demands. Affective experiences—along with distractions and interruptions causing specific affective states—are a core source of attentional demands that interfere with the attentional demands of the primary work tasks. Beal et al. (2005) argued that affective experiences reduce episodic performance because these affective experiences call for affect regulation which, in turn, depletes resources that otherwise could be devoted to the task. A recent empirical study provided support for the core assumptions of this model (Beal, Trougakos, Weiss, & Green, 2006). Specifically, it was found that when experiencing negative emotions persons find it more difficult to regulate their emotions and to follow display rules which, in turn, decreases job performance. More studies are needed that test other facets of the model.
Take away from the changing-subject and changing-task models
Traits, abilities, and skills are changeable over time and as these changing traits, abilities, and skills may interact with the task structure to produce changes in performance. There may be differential rates of change—with some factors changing very slowly even as a result of direct interventions (slow changing traits) and some other changing more quickly (e.g., skills). The rate of change of predictors needs to be determined empirically and should not be assumed to be either all or nothing (Nesselroade, 1991; Srivastava, John, Gosling, & Potter, 2003).
Other info on dynamic performance models
We use the linear model because we want to predict, not because we want to explain how or why a thing is happening a certain way Maybe we don't have the power to always detect the interactions like cubic effects and quadratic Autoregressive predictis itself
Short Term Changes Increasing
What we attend to, can be stored into memory and what we attend to will be rehearsed enough to be kept in memory. Emotional processes are a function of affective events as described above (Weiss & Cropanzano, 1996). Motivational theory is full of descriptions of how motivation can be increased on a short-term basis by changes at work—usually via extrinsic motivation. Often positive reinforcement is seen to produce short-term changes (that are also reduced again, if reinforcements are withdrawn; Eisenberger, Pierce, & Cameron, 1999). Motivational processes may be enhanced too much in the sense of escalation of commitment (Staw & Ross, 1987). While escalation of commitment has positive effects on performance that is the target of motivation (in the sense of working harder, but not in the sense of working smarter), alternative routes and plans for goal achievement may also get rejected or they are not attended to. Thus, a state of over-motivation for one approach can be observed.
Murphy's Stages of Job Performance
When people are in a maintenance stage, they have learned well to perform all major job tasks and are no longer facing situations characterized by novel or unpredictable demands. Murphy (1989) suggested that personality and motivational factors become more important during maintenance stages. When people are in a transitional stage, they are confronted with undefined methods of operation, they must learn new skills and make decisions about topics that are unfamiliar to them (Murphy, 1989). The length of the transitional stage differs between various types of jobs (e.g., it is rather short in traditional assembly line jobs and can last months or even years in more complex jobs) and is also influenced by individual and situational characteristics. In other words, during transitional stages, employees do not yet have routines available for accomplishing their tasks (Frese & Zapf, 1994).