Separating power and speed components of standardized intelligence measures ☆ Eduardo Estrada a , Francisco J. Román b,c , Francisco J. Abad c , Roberto Colom c, ⁎ a Universidad Camilo José Cela, Madrid, Spain b Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, USA c Universidad Autónoma de Madrid, Facultad de Psicología, Madrid, Spain abstract article info Article history: Received 8 July 2016 Received in revised form 19 January 2017 Accepted 4 February 2017 Available online xxxx Gulliksen (1950) established the distinction between pure power and pure speed intelligence tests. Most standardized measures combine power and speed requirements. The speediness component affects test scores' reliability and validity, since it involves variance not due to the mental ability of interest. Here we propose the use of the Stafford's Speediness Quotient (SQ, 1971) for identifying items biased by the speed component. We developed two converging methods based on Structural Equation Modeling (SEM) to assess the validity of the SQ index. The methods concurrently identify items substantially affected by speededness in three standardized fluid intelligence tests with different speed requirements. Basing on the SQ at the item level, a simple strategy for separating the power and speed components of mental ability tests applied under time constraints is pro- posed. This strategy allows an estimation of the respondents' level uncontaminated by the speed unwanted var- iance. This procedure only requires right/wrong responses (e.g., does not need any information external to the test, such as response times) and it is appropriate for medium-small sized samples. A rule of thumb is suggested for identifying items affected by speediness. The simplicity of the proposed procedure allows its use in applied settings for detecting and controlling speed-related variance in tests' scores. © 2017 Elsevier Inc. All rights reserved. Keywords: Measurement Speed Power Test speediness Speediness Quotient SEM 1. Introduction For almost 90 years, researchers have tried to isolate the effect of speed for responding to standardized tests. The first attempts to address this issue date back to the pioneer works by Peak and Boring (1926), Kelley (1927), Spearman (1927), and Baxter (1941). In his handbook—Theory of Mental Tests—Gulliksen (1950) defined a pure speed test as one composed by a large number of items so easy that any examinee would be able to answer properly all of them. The perfor- mance would be measured through the number of items answered in a limited time. On the other hand, a pure power test was defined as one in which all items are attempted and the score depends upon the number of items properly answered. The vast majority of currently administered tests are neither pure power nor pure speed tests, but a mixture of both in unknown propor- tions. Hence, the concept of test speediness is often used to refer to a “situation where the time limits on a standardized test do not allow substantial numbers of examinees to fully consider all test items” (Lu & Sireci, 2007) or to “the extent to which the time limit influences the measurement of examinees' achievement” (Hailey, Callahan, Azano, & Moon, 2012). 1.1. The speediness threat: intelligence is a matter of level It has been claimed that the speed at which examinees answer to a test is, indeed, significantly related to their ability. Thus, for instance, Spearman (1927) assumed that speed and power rely on the same abil- ity. Vernon (1983) supported this view and suggested that the faster an individual is able to process the information, the less time he needs to maintain it on working memory. Hence, working memory capacity will be more efficient combined with a higher speed of processing, lead- ing to a better cognitive performance. However, differences in speed when completing a test might be un- related with power differences in the measured ability. In this regard, Dennis and Evans (1996) showed that scores on a test might depend upon the Speed-Accuracy tradeoff strategy chosen by the person, in- stead of his or her actual power/level. Talento-Miller, Guo, and Han (2013) reported that individuals whose native language was Korean had issues with speediness in many English computerized tests measur- ing cognitive abilities. Consequently, test scores might be affected by the language proficiency of the individuals in some scenarios. Similarly, at least under certain conditions, when individuals are trained in an intel- lectual task, they respond faster and achieve higher performance, re- gardless of their actual level in the measured ability (Chuderski, 2016). Intelligence xxx (2017) xxx–xxx ☆ This research was partially supported by Grant PSI2010-20364 (Ministerio de Ciencia e Innovación, Spain) and by Universidad Camilo José Cela. ⁎ Corresponding author at: Facultad de Psicología, Universidad Autónoma de Madrid, 28049 Madrid, Spain. E-mail address: roberto.colom@uam.es (R. Colom). INTELL-01196; No of Pages 10 http://dx.doi.org/10.1016/j.intell.2017.02.002 0160-2896/© 2017 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Intelligence Please cite this article as: Estrada, E., et al., Separating power and speed components of standardized intelligence measures, Intelligence (2017), http://dx.doi.org/10.1016/j.intell.2017.02.002