I Psych - recall cites Flashcards
Leslie et al., 2014
This is a meta analysis on the stigma surrounding affirmative action policies.
- The presence of AAP influenced others’ perceptions of the beneficiaries’ competence and warmth, which in turn predicted their evaluations of the beneficiaries’ performance.
- When examining self-stigma, the authors demonstrated that beneficiaries were likely to doubt their competence, which led to poorer performance objectively and as evaluated by others.
The Uniform Guidelines debate.
Debate about Uniform Guidelines
- McDaniel et al. (2011) argued the Uniform Guidelines (1974) are scientifically inaccurate and inconsistent with professional practice due to its focus on situational specificity, its lack of discussion of differential validity/prediction and diversity-validity dilemma, and the arbitrariness of the 4/5ths rule.
- Outtz et al. (2011) fought back and said that there is no readily available replacement and the case law generated by the UG would remain. Additionally, there is no need to abolish or revamp the entire document when a) some of it doesn’t need to have its basis in science, and b) some of it has actually been the impetus for science.
Adler et al., 2016
This is an IOP on performance management vs appraisal.
- Arguments against performance appraisal include failed interventions, rater disagreement, weak evaluative criteria, conflicting purposes.
- Arguments for keeping PA include the need for performance mgmt, inherent evaluation of peformance regardless of PA, and merits of ratings
- PM should:
- Enable employees to align their efforts to org goal
- Provide guideposts to monitory behavior and make real-time performance adjustments
- Help employees remove barriers to success
- If managers engaged in effective day-to-day PM behavior as needed, there should be less (if any) need for formal PM systems, such as appraisals/ratings
Campbell et al., 1993
Basic model of work performance
- It’s an expansion of Project A (Army studies), to make it more appropro for non-military jobs.
- Argued that there had been so much focus on predictors but not enough on criterion.
- 8 components of performance, 3 of which are essential part of all jobs: demonstrating core task proficiency, effort, personal discipline. Others may or may not, e.g., management, facilitating team performance, written/oral comms.
- Valuable middle ground approach to viewing performance, rather than being one single broad thing that applies to all orgs, or being something that’s unique in each job. Also helps us focus on what parts of behavior are under employee control.
- Empirically supported
- Was updated in 2012 - revised to be as concrete as possible in defining the 8
Borman & Motowidlo, 1993; 1997
Distinguishes between task & contextual performance behaviors, & presents a taxonomy of contextual performance (containing elements of OCB & prosocial behavior).
- Influenced by Project A and other studies, proposed a model of performance with 2 factors (1993) and 1997 clarified some vague definitions and spelled out 5 OCBs
- Task performance: the proficiency with which job incumbents perform activities that are formally recognized as part of their job
- OCB: go beyond task perf, instead support the org, social and psych context in which work is performed.
- 5 categories of contextual perf: 1) Persisting with enthusiasm & extra effort to complete own tasks; 2) Volunteering for extra-role activities; 3) Helping & cooperating with others 4) Following org rules & procedures; 5) Endorsing, supporting & defending org objectives
- Supervisors consider contextual performance when making overall performance ratings and weight it approximately as highly as task
- Cog ability predicts task perf; Personality predicts contextual performance.
Podsakoff et al., 2000
This is a meta on OCBs.
- OCBs uniquely accounted for much more of performance evaluations than task performance (43% for OCBs; 10% for task). Together they accounted for 62%.
- There are 6 factors that reflect OCB content and they overlap with subfactors of contextual performance.
- Anteced (causes) of OCBs: 4 categories:
- Employee characteristics (i.e., conscientiousness, agreeableness, emotional maturity)
- Task characteristics (i.e., feedback, intrinstically satisfying tasks (job design); role conflict, role ambig (low) - so, jobs can be redesigned to encourage OCBs
- Org characteristics (i.e., org justice, org support, goal setting, org climate) - orgs can change these to encourage OCBs
- Leadership behaviors: communicating vision, role modeling, fostering acceptance of group goals, intellectual stimulation
Types of Behavior in Performance
Task Performance: primary facet- how well someone does their job Org Citizenship (OCB): voluntary helping behaviors that support work context (aka contextual performance, Borman & Motowidlo, 1997; Podsakoff et al., 2000: explains more variance in ratings than task) Adaptive performance: how an employee responses to changes (existing or anticipated) in environment (Pulakos et al., 2000) - i.e., handling work stress, learning new procedures Counterproductive work behaviors (CWB): bad behaviors; orgs try to avoid selecting
4 Appropriate Characteristics of Job Performance Measures
4 appropriate characteristics of job performance measures:
-Individualization - must be data about performance that the individual controls
-Relevance - must be focused on critical parts of the job - i.e., being prompt may not be critical
-Measurability - must be able to generate a # that represents the amount or quality of work performed.
-Variance - Scores must have differences among them to be able to distinguish high/low performers
Gatewood et al., 2019
Forms of Rater Error, how to avoid
4 types of rater error (introduce bias, usually unintentional).
1) Halo: rating the subordinate equally on diff performance items (scales) based on overall impression of worker
2) Leniency: when disproport number of workers get high ratings
3) Severity: when disproport number gets low ratings
4) Central tendency: when a large number get ratings in the middle of the scale
Best way to overcome them is to train supervisors how to avoid them.
Gatewood et al., 2019
Griffin et al., 2007
This is a theory paper with empirical support on a 3x3 model of work role performance.
- Does not focus directly on latent structure of performance.
- Instead posits 3x3 classification of work role behaviors in which one dimension represents org level (indiv, team, org) and the other goes from proficiency on prescribed tasks to proficiency in adapting to changes in indiv, team, or org requirements, to being proactive in instituting new methods or solutions at indiv, team, or org level.
- Three items (i.e., rating scales) assess proficiency within each of the nine cells. the level dimension seems to assess indiv task performance, support in teams, and mgmt role
- proactivity- like OCBs; adaptivity - Pulakos et al., 2000
Heilman & Chen, 2005
This was an empirical study on OCBs and gender.
Men engaging in OCBs were viewed positively. Women displaying the same behavior were seen as simply doing their jobs.
Van Iddekinge, et al., 2017
This is a meta-analysis on interaction of cog ability and motivation on performance.
- Cog ability and motivation are weakly correlated. They are independent, and both relate to performance. This suggests that orgs should measure both variables to predict future job performance. Don’t just emphasize one over the other.
- Additionally, the effects of ability and motivation on performance are additive, rather than multiplicative. This suggests applicants should be allowed to compensate for lower scores on ability with higher scores on motivation scores and vice versa. Instead of multiple hurdles/cutoffs for each, it could be more effective to set a minimum total score for the two measures combined.
- Higher motivation employees could be more impacted by ability training (the interaction was there, just didn’t account for most of variance in performance).
Job Analysis Cites
Gatewood et al. 2019
Job analysis is a purposeful, systematic process for collecting information on the important work-related aspects of a job.
Cascio & Aguinus 2018
- JA is used to define the job in terms of behaviors necessary to perform it and to develop hypoth around what personal characteristics are needed to perform them.
- JA comprised of job description (what the work is) and job specfications (what personal chars are needed to do it.) Specs should include minimally acceptable standards for selection and later performance (i.e., perform at entry? barely acceptable employees be able to do the task/have KSA? importance?)
- Profiles created from method that meet criteria of ratings of description/specs are reviewed and rated on level (of quality of applicant) and clarify to applicants, as well as linkage to the MQs that were created.
- No single type of job-analysis data can support all HR activities. Critical to align method with purpose
- Methods: critical incident, interviews, SME panels, direct observation, questionnaires
- CMs focus on identifying broader characteristics of individuals & using these chars to inform HR practices. Linked more to biz strategy and prescriptive not descriptive. But rigor and documentation of JA is more likely to withstand legal challenge.
DuVernet et al., 2015
-Meta on JA data quality - when rating scales require objective judgements, the data quality is higher. Using multiple methods and training raters did not really increase quality.
Campion et al. 2011
-CM does not inherently lack rigor and can be more manager friendly. Can be used to align all HR systems. CM Best practices: linking CM to org goals, start with top execs, use rigorous JA methods.
Cascio & Aguinus 2018 Job Analysis
This is a chapter on Job Analysis.
- Job analysis is used to define the job in terms of behaviors necessary to perform it (job description) and to develop personal characteristics needed to perform them (specifications)
- JA is used for work/org design, selection, perform appraisal, and other uses
- No single type of job-analysis data can support all HR activities. It is critical to align method with purpose (i.e., interviewing incumbents or having observers; activities or attributes)
- Define the job in terms of both TASK (becomes job description) and PEOPLE requirements
- Elements may include job title, job activities, working conditions/physical envir, social envir, conditions of employment (hours/week, wage, benefits)
- Job specs: KSAOs deemed necessary to perform a job. Bc they exclude people, must set minimally acceptable standards (MQs).
- Job specs based on task and KSA criteria: perform at entry, barely acceptable, importance of correct performance, difficulty.
- Reliability: higher IRR from analysts than incumbents;
- Competency models focus on identifying broader characteristics of individs & using these chars to inform HR practices.
Levine et al., 1997
This refers to a methodology for developing minimal qualifications (MQs).
- Must be set because they exclude some people based on education, experience, etc. For legal defensability purposes.
- A court approved it and it is consistent with sound professional practice.
- Collect info and then create draft list of tasks and KSAs, and get separate groups of SMEs to rate each on a set of 4 scales that indicate the degree to which employees must have/be able to perform the task/KSAO upon entry, the importance to the job that it’s done correctly, and its relative difficulty
- Ratings are aggregated with means or percentages; criteria is set at majority over certain scores for each
- SMEs then provide specific input in terms of # years experience, type of experience, etc.
- Job Analysts then create draft MQ profiles (education, training, work experience to perform the job at satisfactory level).
- NEW set of SMEs then review the MQ profiles to see it if itneeds editing, rate the final profiles on 2 scales, level and clarity, and establish description of barenly acceptable employee
- Those that meet criteria on clarity/level are linked back to tasks/KSAs, to ensure the profile provides an employee with what is needed to perform at barely acceptable level. Those that are linked to more than half of either tasks or KSA or is linked to all 5 most important tasks or KSAs are considered content valid.
Job Analysis Methods
- Direct observation (of incumbents) and performing the job (by analyst). Observation best for manual, standardized, short-cycle activities (i.e., barista). Performing is best for jobs analyst can learn readily. Pros: objective. Cons: assumes jobs are static, can be intrusive.
- Interviews: most common; should be checked for appropriateness in ters of wording and objectivity. Pros: best for reliability, depth of info, immediately clarifies ambiguities. Cons: depends on interviewer skill, distortion of info due to falsification/misunderstanding, time-consuming, social desireability, mistrust.
- SME panels: 6-10 people develop info on job tasks/KSAOs to be used in developing questionnaires and establish links between tasks and KSAOs, test items and KSAOs and tasks and test items. Panels should be representative of work unit and broad level of experiences. Pros: experienced workers provide info, workers able to discuss issues and resolve disagreement. Cons: could be inaccurate (conformity, pressures, etc.)
- Questionnaires: Pros: cheaper, efficient, standardized, easily quantifiable for stats. Cons: $$, time consuming to develop, ambig can’t be resolved real time, no rapport ability
- Fleishman Job Analysis Survey - one of most researched. Describes jobs in terms of abilities required to perform them aiming to list fewest independ categories that describe perform of widest variety of tasks.
- Critical incidents: collection of anecdotes of job behavior (from superv, employees, others) that describe good, bad performance. Each includes what led to performance, what was so effective/not, whether it was under control of employee. Pros: static and dynamic dimensions covered; cons: time consuming to collect and categorize upwards of 100 incidents; difficult to analyze qual data. (Cascio & Aguinis, 2018)
- Other sources: ONET, Training manuals, diaries of work tasks
DuVernet et al., 2015
This is a meta-analysis on job analysis validity and reliability.
- Work analysis data varies as a function of design choices. When rating scales require objective judgements, the data quality is higher.
- Using multiple methods did not really increase quality
- Competency modeling can produce quality data than previously thought (lower IRR but higher discriminability between jobs when analyzing work)
- Training raters only had minimal effects on quality of data.
Campion et al., 2011
This is an article on best practices in competency modeling.
- Competency modeling refers to the collection of KSAOs that are needed for effective performance in the job in question. (The KSAOs are the competencies, and a set of them is a model).
- CM does not inherently lack rigor. But need to use rigorous JA methods.
- CMs are more worker-focused and JA is more job task focused.
- CM is more “manager friendly” in that the KSAOs are usually linked to business objectives & strategies. It’s best to have fewer C’s (8-12) per role; heirarchy can also help.
- CMs used to align HR systems: hire, train, appraise, promote, and pay in terms of the same KSAOs. Those KSAOs are linked to high job performance, biz strategies and goals, and future requirements. Alignment helps promote those goals.
- Best practices include linking CM to org goals, start with top execs, using rigorous JA methods to develop competencies, using both cross-job and job-specific competencies, using comptency libraries; using CM for legal defensability (test validation).
- Unique CM techniques that lend rigor: behavioral event interviews (more in depth than CIs), rating future importance, rating how the C distinguishes from high vs avg perf
Austin & Villanova
This article describes the “criterion problem.”
-
Campbell & Wiernik, 2015
This is a review article on performance appraisal and management. 8 dimensions per Campbell et al., 1993, 1997
- Performance is conceptualized as in-role and extra-role.
- The latent structure of individ work performance is multidimensional, and their 8 factors represent a consensus developed over several decades
- Leadership performance includes: support, initiation, goal emphasis, empowerment, training, and role modeling.
- Management performance includes: problem-solving, goal setting, coordination, monitoring, external representation, staffing, admin, and compliance. (Campbell, 2012)
- Measuring performance must be construct-valid and not tainted by outcomes over which performers have no control. (Campbell, 2012)
- The general factor, “g”, exists in virtually all performance covariance matrices, esp in performance ratings, but is not a single latent variable that can be specified. It must be “formed” as a sum score of diff components for decision making (thru weighting)
DeNisi & Murphy, 2017
This is a review on 100 years of performance mgmt and appraisal in JAP.
-Performance management - wide variety of activities, policies, procedures, and interventions designed to help employees improve performance (i.e., begin with appraisal, but include feedback, goal setting, training, reward systems).
-a formal process, which occurs infrequently, by
which employees are evaluated by some judge (typically a supervisor)
who assesses the employee’s performance along a given set
of dimensions, assigns a score to that assessment, and then usually
informs the employee of his or her formal rating.
Cascio & Aguinus 2018 Performance Mgmt/Appraisal
Performance management: Performance management is a continuous process of: identifying, measuring, and developing the performance of individuals and teams and aligning performance with the strategic goals of the organization. PM serves strategic purpose helps link employee activities with org mission/goals.
Performance appraisal: is the systematic description of job-relevant strengths and weaknesses and is a key component of performance mgmt.
Involves both observation and judgement. Both processes prone to bias.
-Objective data such as production data can avoid bias, but creates other bias, often measure things beyond the employee’s control and measure outcomes of behavior not behavior itself.
-Subjective measures (i.e., supervisory ratings) are prone to other forms of bias.
-Biases can be associated with raters (lack of 1st hand knowledge), ratees (gender, tenure), interaction (race, gender), or situational/organizational factors
-Bias can be reduced sharply through training on both technical and human aspects of the rating process.
Sanchez and Levine, 2009
This is an article on competency modeling.
Reasons to use competency models:
-They’re a solution to companies needing employees to adapt to changing circumstances. Hire people based on adapability and alignment with org mission
-Working and living in a more complex world
-Jobs have been re-defined and more varied than ever
-Results of a JA hold only as long as the job remains the same.
-Job analysis and Competency modeling supplement each other, do not replace each other.
Morgeson et al., 2016
This is an empirical study on job analysis.
- Having too much experience is not good for JA - “too hard to explain”
- This is in favor of data quality checks. Have attention check questions built in, or break up any long surveys - 30 minutes, then break, then 30 more minutes.
- Use this: If you wanted to say something in comps about why you might want to include something beyond task analysis.
- Also, have cutoffs for who takes it (for example experience…because of their findings that people with more experience)
Pulakos et al., 2000
This refers to adaptive performance as performance criteria (study and measurement development).
- Campbell 1999 had suggested that a 9th performance dimension for their 1993 model of performance could be adaptability.
- How an employee responds to changes in the work environment (i.e., change in work procedures, managers, or team”)
- Has 8 dimensions, i.e., handling work stress, solving problems creativity, dealing with unpredictible situations, learning technology and procedures, etc.
Rotundo & Sackett, 2002
This is an experimental study on the relative importance of task, OCB and CWB to ratings of overall performance.
-Although all three components influence ratings of overall job performance,
raters demonstrate unique implicit rating policies that can be grouped into three distinct clusters. The patterns werent’ based on type of job (nursing vs accounting) or org, so likely based more on implicit policies of raters
-task performance weighted highest
-counter-productive weighted highest
-equal and large weights given to task and CWB
They define performance as those actions and behaviors that are under the
control of the individual and contribute to the goals of the organization
Harari et al., 2016
This is a meta on creative and innovative performance.
This study aimed to clarify the distinction between creative-innovative performance (CIP) and other performance dimensions, due to past research showing some overlap.
-Creative-innovative performance is positively related to task performance and OCB, and negatively related to CWB.
-They argue that CIP has emerged as an important component of performance, just as adaptability did. Innovation and creativity are critical to the success of modern orgs, even in orgs that are not focused on introducing new technologies (Zhou, 2008/Zhou & Shaley, 2011)
-Creativity: generation of new ideas; Innovation: new ideas + implementation
-CIP refers to the proficiency with which employees generate and implement novel ideas in the workplace - both outcomes and the behaviors are included in conceptualization. Predictors of this would be the processes.
Meinecke et al., 2017
This is an empirical study on manager-employee communication during performance appraisals.
-Employee and supervisor perceptions of performance appraisal meetings/interviews were positively impacted by a communication pattern characterized by relational communication followed by employee active participation.
-These patterns were linked to higher interview success ratings by both supervisors and
employees.
Levy & Williams, 2004
This is a review and framework on the importance of the social context in the performance appraisal process.
- Takeaway: PA literature is more cognizant of the importance of the social context
- Distal factors: i.e. org culture/climate; economic conditions, legal climate, HR strategies, Org goals and others.
- Process Proximal variables: have direct impact on how appraisal process is conducted (i.e., rater issues - affect, motivation, accountability; ratee issues - how PA affects motivation; LMX - higher LMX leads to higher ratings (mixed results); politics, feedback environment
- Structural Proximal variables: have direct effects on rater and ratee behavior; directly affected by distal (i.e., multi-source feedback systems; performance appraisal purpose (ratings more lenient when used for administrative purposes than for developmental); rater training
Motro and Ellis, 2017
This is an empirical study of feedback responses and gender.
- Men who cried in response to negative feedback were seen as atypical and were rated lower on performance and leadership capability. The same was not true for women.
- This is explained by role congruity theory (Eagly)
- Bias can affect feedback providers and their evaluations of employees.
Alliger et al., 1997
This is a meta-analysis on training criteria (outcomes that should be evaluated).
- Expanded Kirkpatrick’s (1994) taxonomy of training criteria.
- Reactions split into affective reactions and utility judgements
- Learning split into behavior/skill demonstration (during training), immediate knowledge (right after training), knowledge retention (over longer period of time after)
- Behavior renamed as transfer, which is a useful clarification
- Results stayed as “results” (org level results)
Huang et al., 2017
This is a longitudinal study on training transfer.
- The degree to which training transfers for an individual is dependent on motivation to transfer.
- Initial transfer attempts are dependent on post-training self-efficacy
Keith and Frese, 2008
This is a meta of error management training.
- EMT leads to better outcomes than non-error management training.
- EMT leads to better transfer than non-EMT, especially for adaptive transfer (transfer to novel tasks).
- EMT may be better suited than error-avoidant trainning methods for promotion of transfer to novel tasks (useful in today’s rapidly changing work environment)
Colquitt et al., 2000
This is a meta-analysis of training motivation.
- Training Motivation predictors: individual (internal LoC, achievement motivation, conscientiousness, anxiety, pre-training SE, supervisor and peer support, job involvement, org commitment) and situational (climate)
- Training Motivation predicted: knowledge, skill acquisition, transfer, reactions, post-training SE
- Cog ability mediated the effects on learning outcomes. But training motivation explained incremental variane beyond cog ability.
Salas et al., 2012
This is a review on practical training.
- Pre-training needs: perceived support, motivation, anticipated opportunity to use skills
- Needs analysis: job, organizational, and person analyses
- Good learning climate requires realistic expectations, prep, training framed as an opportunity, and reinforcement
- Individual differences: SE, goal orientation (learning), motivation
- Key features of training: information, demonstration, practice, feedback
- Transfer influences: transfer climate, post-training environment, supervisors, debriefing (refresh), communities of practice, opportunities to use learnings
- Training evals: precise ABC (affect, behavior, cognit) measures that reflect outcomes and tie to evaluation purpose.
Blume et al., 2010
This is a meta of training transfer.
- Open (generalizeable) skills are more transferrable than closed (single-use) skills
- Strongest predictors of transfer: cog ability, conscientiousness, supervisor and peer support, post-training SE, utility reactions.
- Transfer is highest immediately after training and requires maintenance training to avoid skill decay.
