min read

Effective hiring – how it exists in personnel selection measures

Examining what some of the best selection measures are for effectively hiring new employees.

By: Cathy Xie

Examining the Validity of Selection Measures

Ever wonder what the best selection methods are when hiring new employees? In this blog post, we will review an article by Schmidt & Hunter (1998) which can help you determine exactly that. The researchers conducted a meta-analysis (a statistical method that allows you to combine the results of multiple studies to examine their effects; Hedges, 1992) to examine 85 years of research in personnel psychology. They specifically examined the validity of 19 types of assessments that are currently used to select and hire potential employees as well as measure the impact of training and developmental initiatives. The researchers investigated the effectiveness of each of these measures individually and combined with other methods of selection. This blog can help you determine which assessments contribute to effective hiring and help you to increase your workforce’s productivity and learning ability.  

Key Findings

Schmidt and Hunter (1998) found that the measure with the highest validity in predicting job performance was work sample tests. Work samples are assessments that measure an individual’s skills or abilities by examining their behaviour when completing a task (or set of tasks) that is comparable and relevant to the tasks they would perform on the job (Robertson & Kandola, 1982). This means, of the methods reviewed in this article, that work samples are the best indicator of how someone will perform on the job (for more information about forms of validity, check out this blog post). Following closely behind work samples are tests of general cognitive ability/general intelligence (GMA) and structured employment interviews (e.g., interviews that use a standardized set of questions for all candidates and using standardized rating scales). Furthermore, Schmidt and Hunter (1998) found that assessments such as graphology (the way someone writes), and information about one’s years of education, interests and age have low predictive validity. That is, these assessments such as these have little to no ability to predict job performance.

When it comes to on-the-job learning, Schmidt and Hunter (1998) found that GMA is the strongest predictor of one’s success in training programs, followed by integrity tests (a measure of one’s trustworthiness and dependability – often examining the likelihood that employees will engage in negative behaviours like stealing from employers and drinking on the job), peer ratings of the candidate’s job performance, and employment interviews. In other words, when employers choose job candidates using their GMA, they are choosing candidates who will learn the most from training sessions and be able to acquire the most knowledge on the job. It should be noted, that work sample assessments were not examined when determining their effectiveness to predict success in on-the-job training.  

Next, the Schmidt and Hunter (1998) examined the increased validity of measures when they are combined with GMA. Among the 18 measures, integrity tests combined with GMA demonstrated the highest predictive validity compared to the other types of selection measures. This demonstrates that assessments of one’s GMA and integrity both provide unique pieces of information about a candidate, allowing you to better predict job performance when they are both used during the hiring process. Other methods with comparable levels of predictive validity when combined with GMA are work sample tests and structured employment interviews. This article does not examine the combinations of assessments beyond GMA, though more recent articles have conducted studies to examine this (e.g., job-relevant simulations provide some unique information above and beyond an assessment of one’s work relevant knowledge, when predicting how someone will perform on the job; Lievens & Patterson, 2011).

It is important for us to note, that although GMA is a good predictor of performance and on-the-job training, recent research has found that scores on measures of one’s intelligence/general mental ability can have an adverse impact on different ethnic groups (discriminatory bias against a protected group as a result of a seemingly neutral company practice; Outtz & Newman, 2010). The differences in scores between groups has been found to be unfairly influenced by an individual’s characteristics such as their language fluency and whether they are a first- or second-generation immigrant (Hausdorf & Robie, 2018). Potential impacts of different measures should be considered when deciding the kinds of hiring tools to use in your hiring process.

How Do These Findings Relate to nugget.ai’s Work?

Similarly, to the study conducted by Schmidt & Hunter (1998), we, at nugget.ai, seek to develop measures based on valid empirical research and improve the effectiveness of hiring for employers. Our assessments were inspired by work sample assessments, the assessment demonstrated to be most accurate when predicting performance on the job (to dive a bit deeper in our assessments, check out this blog). We also advocate for a holistic approach when selecting for potential employees. Although hiring is possible by utilizing one type of selection measure, as demonstrated by this article, the most effective hiring is done by utilizing several measures that allow you to understand who a candidate is more comprehensively.

screenshot of 360 candidate view in nugget.ai platform
360-view of candidate skills

Curious about our tools and what to consider them for use in your selection? Contact us here to get the process started!


Hausdorf, P. A., & Robie, C. (2018). Understanding subgroup differences with general mental ability tests in employment selection: Exploring socio‐cultural factors across inter‐generational groups. International Journal of Selection and Assessment, 26, 176-190. http://doi.org/10.1111/ijsa.12226

Hedges, L. V. (1992). Meta-analysis. Journal of Educational and Behavioural Statistics,, 17(4), 279-296. http://doi.org/10.3102%2F10769986017004279

Lievens, F., & Patterson, F. (2011). The validity and incremental validity of knowledge tests, low-fidelity simulations, and high-fidelity simulations for predicting job performance in advanced-level high-stakes selection. Journal of Applied Psychology, 96(5), 927-940. http://doi.org/10.1037/a0023496

Roth, P. L., Bevier, C. A., Bobko, P., Switzer, F. S., & Tyler, P. (2001). Ethnic
group differences in general mental ability in employment and educational settings: A meta‐analysis. Personnel Psychology, 54, 297–330. http://doi.org/10.1111/j.1744‐6570.2001.tb00094.x

Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124(2), 262-274. http://doi.org/10.1037/0033-2909.124.2.262