当前位置：首頁 > 人力资源管理 > 正文内容

《国际人力资源管理》课程公共资源 > 讲义：section08

clarke1周前 (09-10)人力资源管理65

Selection is the process of making a “hire” or “no hire” decision regarding each applicant for a job, which is a process for organizations to choose its members.

A programmer recalled his experience of six hours interview with six separate interviewers in Microsoft, only in the last interview he was asked to write program codes, and the interviewer did not ask him to introduce himself. Needless to say, he was so nervous which may affect his performance during the whole interview.

Any organization must take the utmost care with how it chooses organizational members, however, companies have lots of problems in employee selection.

Therefore, several generic standards should be met in any selection process. We focus on five: (1) reliability, (2) validity, (3) generalizability, (4) utility, and (5) legality.

Reliability is the degree to which a measure of physical or cognitive abilities, or traits(such as intelligence), is free from random error.

The reliability of the computer is very important.

There are three types of reliability estimate:

1. Test-retest estimate. The same test was carried out on the same object at two different time periods.

2. Equivalent-form estimate. Compare two tests, the second test is believed to be equivalent with the previous one by experts.

3. Internal comparison estimate. Using statistical methods to analyze the internal consistency of the test.

In statistics, the correlation coefficient is used, which is a measure of the degree to which two sets of numbers are related.

A perfect positive relationship equals one.

A perfect negative relationship equals one below zero.

Validity is the extent to which performance on the measure is associated with performance on the job， which means that employees who do well in the test will be more outstanding in the work, and those who don’t do well in the test will have relatively poor performance in the job..

Validity is used to prove that the test is related to the work, and the test is an effective predictor of the performance of work.

The types of validity included predictive validation, concurrent validation and content validation.

In predictive validation, the company measures all job applicants on attribute, then hire some applicants and reject others. After some time period, the company measure all newly hired incumbents’ performance, then the correlation between these two sets of numbers can be obtained.

In concurrent validation, the company measures all existing employees on attribute, and their job performance, then the correlation between these two sets of numbers is obtained.

The predictive validation compare two sets of test numbers over time, while concurrent validation compares two sets of numbers at the same time.

Please read table on the slide, what is the relationship between the required level of correlation and sample size?

The bigger the sample is, the smaller the required correlation is.

So criterion-related validation is not appropriate for small companies.

The ability to use content validation in small sample settings makes it generally more applicable than criterion-related validation.

Content validation is performed by demonstrating that the items, questions, or problems posed by the test are a representative sample of the kinds of situations or problems that occur on the job.

As the slide shows, if one construction company recruit typist, the job applicants are supposed to have a typing test using sample A or sample B.

Which sample is considered to have content validation?

If the content of the test is a representative sample of the work, it has content validation.

Suppose two college graduates apply for the typist in the construction company, one of them has internship experience in other construction companies.

So do you think the testing performance guarantees the one who has internship experience is more qualified than the other one who has none? Not necessarily.

However, content validation has limitations.

Furthermore, subject judgment plays a large role in the content validation.

Generalizability is the degree to which the validity of a selection method established in one context extends to other contexts, including different situations, different samples of people, and different time periods.

For example, The SAT is a standardized test widely used for college admissions in the United States.

Rather than rely on the SAT for all types of undergraduate programs, separate tests like the GRE and GMAT may be needed for particular types of graduate schools.

GRE(Graduate Record Examination) is postgraduate entrance examination in USA.

GMAT(Graduate Management Admission Test) is usually required by business schools.

Utility is the degree to which the information provided by selection techniques enhances the effectiveness of selecting personnel in organizations.

In general, the more reliable, valid, and generalizable the selection method is, the more utility it will have.

In addition to being influenced by reliability, validity, and generalizability, the utility is influenced by the value of the products or services produced by workers, as well as the impact of the cost of testing.

Suppose one selection test is highly reliable and valid, but the cost is very high, if you are the manager, will you use such test in your employee selection?

In classical view, management’s only social responsibility is to maximize profits.

The final standard that any selection method should adhere to is legality.

All selection methods should conform to existing legal precedents. Employers who are taken to court for illegal discrimination experience high costs with litigation, and also suffer potential damage to their social reputations as good employers, making recruitment more difficult.

Here is a news report for Walmart.

1.5 million current and former female

employing illegal aliens, discriminating against women and people with disabilities

Wal-Mart is the world’s largest retailer, employing more than 1.5 million people worldwide. But it has been criticized for the way it treats these workers.

Therefore, five generic standards should be met in any selection process: (1) reliability, (2) validity, (3) generalizability, (4) utility, and (5) legality.

For example, a measure must be reliable if it is to have any validity. On the other hand, we can reliably measure many characteristics (like height) that may have no relationship to whether someone can perform a job in most cases.

For this reason, reliability is a necessary but insufficient condition for validity.

A measure must be reliable, valid and generalizable, if it is to have any utility.

The first four build off each other in the sense that the preceding standard is often necessary but not sufficient for the one that follows.

This is less the case with legal standards. However, a thorough understanding of the first four standards helps us understand the rationale underlying many legal standards.

B c