Ch. 7 I:O Psychology- Selection Decisions and Personnel Law PDF

Title Ch. 7 I:O Psychology- Selection Decisions and Personnel Law
Course Industrial/Organization Psychology
Institution University of Akron
Pages 8
File Size 125.7 KB
File Type PDF
Total Downloads 6
Total Views 146

Summary

These are all the notes from in-class lectures on Chapter Seven....


Description

Ch. 7 I/O Psychology: Selection Decisions and Personnel Law Overview  Basics of selection Batteries  Making Decisions  Utility  Legal issues in selection Use selection batteries to make hiring decisions  Set of tests or predictors  Predictors based on Job Analysis  Effective batteries predict success better than use of any one test alone. Selection Battery:

Work Sample 2

Assessment r 2=¿ Center 0 25

r =¿ 0 36

r 2=¿

Criterion Performance

The sum of the r 2 from the circle doesn’t equal the actual r 2 of the selection battery due to the overlap. However, they all add knew validity.

Interviews

Selection Decisions  Should be based on valid tests/predictors  There are two approaches to demonstrating validity of a selection battery o Local validation study (predictive or concurrent) o Validity generalization Local Validation Study  Predictive: test scores obtained at one time point will predict criteria obtained at some future time point Steps:





1. Gather predictor data on all applicants 2. Hire applicants based on predictors not part of the selection battery 3. Gather performance data to serve as criteria for validation 4. Computer between predictor and criterion scores Concurrent Validity: How well a test predicts a criterion that is measures at the same time as the test o Often more viable than predictive approach Steps: 1. Collect data on both predictors and criteria from incumbent employees at the same time 2. Compute r to assess strength of relationship between predictor and criterion Difference between predictive and concurrent o Concurrent uses people already working at a job (incumbents) while predictive looks at applicants, then them as incumbents o Predictive (T1 to T2); Concurrent (T1 to T1)

Situational specify  What if another organization already conducted a validation study?  Do different organizations need to demonstrate validity using their own study? o At first they said, yes, absolutely, due to situational specificity Validity Generalization  VG challenges assumption of situational specificity  Use meta-analysis to weight and combine r’s across situations o Examine the generalizability of r’s o Research had supported the notion that validities do generalize  VG Hang-ups o Many criticism and concerns about the statistical methods used in VG o Limited to jobs very similar to the jobs on which test was originally validated o Courts have some concerns with this approach, but other times the approaches held up in court when it’s done carefully and a JA was used to show jobs are the same Rehash 1. Choose valid test based on JA a. Determine validity via local validation or VG 2. Applicants take tests 3. You have applicant test scores, now what? 4. Need to make hiring decisions Making Decisions Overview  Approaches to hiring decisions o Multiple cutoff

 Multiple Hurdle o Multiple regression Multiple Cutoff  Non-compensatory approach o There will be “passing scores” or cutoffs for each predictor o Applicant has to score higher than cutoff on each predictor  This means if you are the best in four areas, but you don’t meet the cutoff in the 5th area, you don’t get hired  Strengths: Doesn’t allow candidates weak in very important areas of the job to be selected despite other abilities (ex. Surgeon)  Multiple Hurdle: o Special form of multiple cut off o Predictors administered in specific order o Only those that pass a test or hurdle move onto the next one o More cost and time effective Multiple Regression  Statistical technique to estimate how well a set of predictors forecast criterion o Generates equation that weights each predictor  It is a compensatory approach (high in one area can make up a low score in another area)  Regression equation: y-predicted = intercept + weight (x1)  Considerations o May still want to set a cutoff on the predicted criterion o May still want to set a minimum cutoff on each predictor  Sort of hybrid approach Usefulness of selection process  Is it worth it? o A question of utility  Element of selection process for assessing utility: 1. Decision accuracy 2. Validity 3. Base rate 4. Selection ratio 5. Cost 1. Decision accuracy:  Whole point is to only hire people who will be successful and reject those most likely to fail  System needs to be accurate



Possible outcomes:

4: misses

1: hits

Success

Failure

2: correct rejections

3: false alarms

Reject Hire *People will be speckled throughout the entire graph and in every quadrant*  1 and 2 are good decisions  3 and 4 are bad decisions 1. Decision accuracy continued:  Percentage of hiring decisions that were correct o Not looking at people not hired only at people that were hired o Quadrant 1 and 3 o Formula: Q1/ (Q1 + Q3) o So hits divided by all hired, which are hits and false alarms  Very important to have high percentage with “dangerous jobs” ex. surgeons, cops, pilots, and etc. Overall accuracy:  Maximize hits and correct rejections (Q1 + Q2)  Minimize misses and false alarms (Q3 + Q4)  So, formula is: (Q1 + Q2)/ (Q1 + Q2 + Q3 + Q4) 2. Validity  Valid selection batteries maximize hits and correct rejections while minimize misses and false alarms  Predictors with larger r’s forecast subsequent performance better  High validity would be closer to a line within the graph above (going through Q1 and Q2)  Low validity would not be a line and the graph above, dots would just be everywhere 3. Base Rate  Percentage of current successful employees on the job  Reflects quality of previous selection batteries  Provides a basis of comparison for new battery

4. Selection Ratio  Number of job openings divided by number of applicants

o If 20 openings and 20 applicants, then SR = 1.00 o If 5 openings and 20 applicants, then SR = 0.25 o Smaller SR means more utility 5. Cost  Development and implementation of a selection battery = $$$  Balance cost and benefit  High stakes selections o NASA and FBI Legal issues in industrial psychology  Before 1960’s, not illegal to discriminate when hiring  Changed with important federal legislation and guidelines o Title VII of the civil rights act (1964) o Uniform guidelines on employee selection procedures (1978) o Principles for the validation and use of selection procedures (2003) Employment at Will  Employers and employees have the right to initiate and terminate the employment relationship at any time, for any or no reason at all  Due to recent employment cases about wrongful discharge, many companies (and states) have implemented JUST CAUSE discharge policies o Discrimination o Breach of contract o Breach of implied covenant of good faith o Violation of public policy  Just cause: Firing someone for acceptable reasons (reasonable) Adverse Impact  Legal operationalization of discrimination  Legalization recognizes two cases of adverse impact 1. Disparate impact: unintentional 2. Disparate treatment: intentional  Calculating adverse impact 1. Rule of thumb: a selection procedure has an adverse impact against a group if the selection rate for that group is less than 80% of the selection rate for the group with the highest selection rate  4/5th rule  Key: AI at any stage is enough for a complaint and “intent” is irrelevant  Steps in a discrimination case 1. Plaintiff demonstrates a prima facie case (on the face of it) 2. Defendant can argue against statistics by showing the plaintiff is only looking at partial data, interpreting it incorrectly, or looking at the wrong data 3. OR, if AI does exist, the defendant can combat the charge of illegal discrimination



  



Although AI does exist, the selection battery is job-related for all minority group and the majority group “business necessity”  Supported with local validation or validity generalization Plaintiff could show that another equally valid test is available resulting in lower AI Plaintiff might try to show that this was all a pretext for discrimination Defendant could claim that it is absolutely necessary to hire people of a particular group because the business cannot be adequately conducted without them… Bona Fide Occupational Qualification (BFOQ) BFOQs can only apply to sex, religion, or national origins under title VII. (This does not include race nor color)  Ex. of males as BFOQs: actors, security guards, and attendants in male restrooms  Ex. of female BFOQs: actresses, lingerie sales person, custodian in women’s dormitories  Court Case Examples of BFOQs: o Pan am Airlines  They did not want to let a male be a flight attendant, the company lost o Hooters  They want to hire female waitresses and will mostly settle disputes out of court. They have also tried calling their waitresses actresses to get away with not hiring males

Affirmative Action  Organizational practice to increase the number of minorities or protected class members or protected class members in targeted jobs  Purpose is to address historical discrimination not a quota system  A.A. Case Law examples o Some programs were popular for many years and also very controversial o Gratz v. Bollinger (2003) - Gratz won because they were giving non sense points to minority applicants o Grutter v. Bollinger (2003) – Grutter lost because the admissions board wasn’t blatantly giving minorities an upper hand. o Both cases were over white applicants not getting into programs when they were qualified for them

Overview: Important Employment Laws  Equal Pay Act (1963)  Civil Rights Act (1964, 1991)  Age Discrimination Employment Act (1967)

 

Americans with Disabilities Act (1990) Family and Medical Leave Act (1993)

Equal Pay Act (1963)  Illegal to discriminate in pay and benefits on the basis of sex, for jobs that are equal o Ledbetter v. Good year Tire and Rubber Co. (2007)  She lost the case because she had 300 days after employment to notice the discrimination  Then, Lilly Ledbetter Fair Pay Act (2009)  You now have 300 days after each act of discrimination to file a lawsuit Civil Rights Act (1964, 1991)  Prohibits employment discrimination based on race, color, religion, sex, or national origin  Enforced by EEOC  Amended in 1991 to clarify monetary damages, jury trials, and party obligations  Griggs v. Duke Power (1971) – African Americans were not being promoted based on an ACT like test o Distinguished DI and DT  Even if no intention to discriminate, if there is AI, it must show job relatedness of selection battery  Ricci v. Destafano (2009) – not acceptable to throw out a valid test because of potential AI  CRA and sexual harassment o Sexual harassment is prohibited under CRA  Quid Pro Quo – This for that  Hostile work environment o Zarda v. Altitude Express (2018)  Man said he was fired because he was gay  Executive Order 11246 o Prohibits discrimination by employers holding federal contacts or subcontracts Age Discrimination in Employment Act (1967)  Prohibits discrimination against people 40 years of age or older  Age as a BFOQ sometimes holds up in court  In discharge cases, plaintiff must demonstrate that they: o Are a protected class o Were doing satisfactory work o Discharged despite work performance o Were replaced by a younger person  Cleverly v. Western Electric (1979) o Cleverly won because he could prove all of the conditions mentioned previously

Americans with Disabilities Act (1990)  Prohibits discrimination against qualified individuals with disabilities in employment decisions  Qualified individual: those who can perform the essential functions of the job  Essential functions: tasks that are significant and meaningful to the job  Employers must make reasonable accommodations for such individuals  Reasonable accommodations: changes in job function that allow qualified, disabled individuals to successfully perform the job  Undue hardship: employers are only required to make reasonable accommodations to the extent that such accommodations do not impose an undue hardship with respect to difficulty of implementation or expense  Job analysis are very important to define the scope of jobs and determine what kinds of accommodations can be made Family and Medical Leave Act (1993)  Allows eligible employees to take job-protected, unpaid leave for up to 12 weeks because of family related issues like the birth of a child, serious health condition of family member, or one’s own serious health condition  Created to help employees balance work pressure with family pressures  Right to return to the same position or equivalent position with same pay and benefits...


Similar Free PDFs