Applicant Reactions to a Situational Judgment Test used for Selection into Initial Teacher Training

We considered applicants’ perceptions of the use of a pilot situational judgment test (SJT) designed for selection into primary and secondary teacher training programs in the UK. Quantitative and qualitative data were collected from 304 applicants (73% female) to two postgraduate (PGCE) training programs in the 2013-2014 application cycle. Participants were asked to provide feedback on the content of the SJTs and on the appropriateness of the tests for selection into teacher training. Results from the rating scales showed that most of the applicants (76.7%) found the content and format of the pilot selection tool favourable. Results from open-ended questions suggested that applicants were aware of issues of procedural justice and fairness in selection for teacher training, with a recommendation that separate selection tests should be created for primary and secondary applicants.

Palabras clave: selección del profesorado, pruebas de juicio situacional, educación del profesorado, atributos no-cognitivos  lthough the claim is sometimes made that effective teachers are made, not born (e.g., Marzano, Frontier, & Livingston, 2011), systematic differences in teaching effectiveness within cohorts of new teachers emerge early and remain intact through at least the first five years of teaching (Atteberry, Loeb, & Wyckoff, 2013).Teacher training providers need accurate, reliable, and evidence-based methods to select prospective teachers because effective teachers make a long-term impact on student academic achievement and well-being (Hanushek & Rivkin, 2012).The current article describes applicant reactions to a SJT that was used in the selection process for primary and secondary teacher training programs in the UK.In this study, applicants for teacher training programs completed a 35-item situational judgment test (SJT)-a scenario-based measurement method designed to assess individuals' judgment in contextualized workplace settings (e.g., Ryan & Ployhart, 2014)-designed to measure the non-cognitive 1 attributes of prospective teachers.
Our focus in this article is on applicants' perception of the relevance and fairness of the use of SJTs as part of the selection process for entry into teacher training.We asked, What are teacher trainees' reactions to the SJT as a selection tool for entry into primary and secondary teacher training programs?Using a theoretical framework of organizational justice (e.g., Gilliland, 1993;Patterson, Zibarras, Carr, Irish, & Gregory, 2011), we examined applicants' reactions to the process of completing a pilot SJT for selection into teacher training.The results of the study provide a starting point for research that considers the selection process into teacher training from the applicants' perspective, and allows researchers and training providers to consider the perceived and actual fairness of selection procedures.

Selection of Candidates for Teacher Training
1 We use the phrase 'non-cognitive attributes' to refer to non-academic professional attributes such as resilience, empathy, teamwork, and integrity.

IJEP -International Journal of Educational Psychology, 3(2) 107
Education systems need valid selection procedures because selecting the right people to work as teachers is critical for a nation's educational, social, and economic well-being (Hanushek & Rivkin, 2012).Unfortunately, teacher selection (and selection for teacher training) practices are often ineffective and unsystematic (Barber & Mourshed, 2007;Staiger & Rockoff, 2010;House of Commons Education Committee, 2012), having received little attention from educational psychology researchers (e.g., Rutledge, Harris, Thompson, & Ingle, 2008), who in turn have been accused of 'irrelevance' with regards to education practice (Berliner, 2006).Policymakers and practitioners have started to question the reliability and validity of the existing selection procedures, particularly since the selection process is time-consuming, resource-intensive, and dependent on the skill and intuition of the interviewers.For example, in April 2014, the Australian Government's Teacher Education Ministerial Advisory Group announced teacher quality as a top priority.By focusing on measurements that identify those best suited to the teaching profession, the advisory group aims to discover valid and reliable measures that address the importance of applicants' numeracy and literacy, as well as interpersonal skills and aptitude for teaching (Commonwealth of Australia, 2014).The quest to improve selection practices for teacher training is seen as a first step in improving teaching quality in a range of international contexts.
Identifying the necessary cognitive dimensions (e.g., subject area knowledge, reasoning ability, and literacy and numeracy skills) of prospective teachers is relatively straightforward, with school and university records, and numerous cognitive reasoning instruments widely available.However, assessing the essential non-cognitive attributes-interpersonal skills, motivational tendencies, and personality traits-shown to be crucial for successful teaching (Rimm-Kaufman & Hamre, 2010), is much more challenging to measure reliably.We are unaware of data on the predictive validity of prospective teacher selection practices, and we know little about the nature of what is being assessed, or how systematic these assessments might be.Calls to improve the teacher training candidate selection process and to identify these non-cognitive attributes have been made by national (e.g., House of Commons Education Committee, 2012) and international (e.g., OECD, 2005) education authorities, and have long been the holy grail of teacher selection research (Barr, 1952).In the UK alone, there are about 108 Klassen et al. -Applicant reactions 125,000 applications under consideration (April, 2014) with about 25,000 placements (Universities and Colleges Admissions Service, 2014), meaning that even modest improvements in the prediction of potential teacher effectiveness have the potential to show substantial long-term benefits for student outcomes (Hanushek & Rivkin, 2012).
Selecting individuals for teacher training is essentially a predictive research exercise, but current selection procedures have not benefited from a strong research base (Rutledge et al., 2008).However, recent advances in educational and organizational psychology provide new ways to understand teachers' non-cognitive attributes and selection procedures.Three recent research advances have changed the teacher candidate selection landscape.First, research on non-cognitive attributes of effective teachers has advanced substantially in the last decade, with educational psychology research focused on the psychological characteristics of effective teaching, such as tacit knowledge (Elliott, Stemler, Sternber, Grigorenko, & Hoffoman, 2011), teacher-student relatedness (Allen et al., 2013), and resilience (Beltman, Mansfield, & Price, 2011).In short, we know more about the non-cognitive attributes of successful teachers than ever before.Second, methodological advances in selection procedures in business settings have resulted in valid and reliable implicit approaches (as opposed to explicit or self-report approaches) for evaluating key psychological characteristics related to teaching effectiveness (Motowidlo & Beier, 2010;Schultheiss & Köllner, 2014).The emergence of SJTs present an evidence-based and innovative approach to assessing non-cognitive attributes of prospective teachers.Third, selection research in other professional fields, such as medical and dental training in the UK, shows robust predictive and incremental validities using SJTs, with direct relevance and application for selection into teacher training (e.g., Patterson, Ashworth, Mehra, & Falcon, 2012).

Situational Judgment Tests and Selection
In conventional selection procedures, non-cognitive attributes are typically assessed explicitly through self-report personality tests or interviews, for example by asking, "Are you good at encouraging others?"The problem with self-reports in high-stakes situations is that respondents can potentially 'fake' their responses and portray themselves in the best possible light.By using knowledge-based instructions, in contrast to behavioural-based instructions (i.e., "What should you do?" rather than "What would you do?") in SJTs, Whetzel & McDaniel (2009) suggest that faking can be reduced.Knowledge-based instructions allow for the assessment of whether the candidate knows the best response to the situation.
In contrast, SJTs can measure implicit traits and attributes.Motowidlo, Hooper, and Jackson (2006) propose the concept of implicit trait policy (ITP), whereby psychological traits can be indirectly evaluated by asking an individual to judge the effectiveness of responses to situations designed to elicit targeted traits.Implicit trait policies are formed through a combination of a person's experience and personal dispositions (e.g., inter-personal skills, motivational tendencies, and personality traits).In addition, theories of teachers' tacit knowledge (e.g., Elliott et al., 2011;Grigorenko, Sternberg, & Strauss, 2006;Stemler, Elliot, Grigorenko, & Sternberg, 2006) suggest that teachers' non-cognitive attributes may be learned implicitly though general socialization and life experiences (such as through early parental modelling) even before professional training commences, and are not transmitted directly through instruction.
SJTs can be used as one part of a selection system to complement a range of other information-rich sources, including tests of literacy and numeracy skills, reviews of academic background, in addition to individual or group interviews.A significant advantage of SJTs compared to other selection tools is that a large range of non-academic attributes can be tested reliably and efficiently as SJTs are machine-markable.SJTs are complementary to interviews as in interviews, although relatively resource intensive, candidate responses can be probed to elicit further relevant information.In selection for medicine in the UK and Belgium, for example, SJTs do not replace other selection processes, but rather complement other tools such as cognitive ability tests and interviews (Patterson, Tavabie et al., 2013).A further advantage of using SJTs for screening is that they show high levels of predictive and incremental validity, fewer inter-group differences (e.g., racial bias; Shultz & Zedeck, 2012), and are more economically feasible than other selection tools such as interviews of cognitive ability tests used for initial screening (Patterson, Tavabie et al., 2013).The placement of SJTs for selection depends on context-specific selection processes.In selection for medical training in the UK and Belgium, SJTs are used as one tool to screen 110 Klassen et al. -Applicant reactions applicants for training placements, and were shown to be effective predictors of subsequent job performance with corrected validities from r = .30to r = .70(Patterson et al., 2012).Using SJTs for selection for teacher training may offer advantages in comparison to other selection tools, but to this point little research has explored how applicants might perceive the use of SJTs as part of the selection process.

Building a Pilot SJT for Teacher Selection
We followed the work of Patterson and colleagues (e.g., Patterson, Lievens, Kerrin, Munro, & Irish, 2013)

Organizational Justice Theory and Applicant Reactions to Selection Procedures
Applicants' perceptions of fairness, feasibility, and reasonableness of selection processes are important for recruitment, ethical, and legal reasons (Gilliland, 1993).In addition, candidates' perceptions of the selection process influence their attraction to the organization (Walker et al., 2013).From a recruitment perspective, a teacher training program's ability to successfully recruit applicants is influenced by the perceptions of current and past applicants, who may share word-of-mouth accounts about the fairness of the selection process, ultimately influencing the success of recruiting the best possible candidates.From an ethical perspective, selection into teacher training must both be fair to applicants, and be perceived as fair by applicants.From a legal perspective, selection processes must not discriminate based on applicants' personal characteristics (e.g., ethnicity, socio-economic background).
Organizational justice theory refers to the perceived fairness of the selection procedures (procedural justice) and the fairness of the selection outcomes (distributive justice) (Hausknecht, Day, & Thomas, 2004;Patterson et al., 2011).The perceived fairness of recruitment activities is important for applicants' well-being but can also strongly influence future recruitment activities and organizational health.Patterson et al. (2011) suggest that applicants' perceptions of fairness are influenced by test characteristics (i.e., the qualities of the testing procedures) and by interpersonal treatment, that is, applicants' perceptions of how they are treated during the selection process.These overall perceptions of fairness about the selection process and selection outcomes influence candidates' decision-making about continuing on in the selection process, whether or not they will re-apply if unsuccessful, and whether they will accept a training 112 Klassen et al. -Applicant reactions place if offered (Patterson et al., 2011).For teacher training institutions, and especially institutions that are operating in a competitive environment where applicants may have several options for training, the perceived fairness of the process is an important consideration.
In Figure 1 we present a model that shows how SJTs influence applicants' personal characteristics and perceptions of procedural characteristics, both of which feed into self-perceptions (e.g., self-efficacy about interview performance).The personal characteristics of interpersonal skills, knowledge and work experience, and personality traits interact with perceptions of procedural characteristics such as interpersonal context, formal characteristics, and selection information.The interaction of personal and procedural characteristics informs applicants' perceptions of overall fairness, which in turn influences explicit reactions about the entire selection process.Our model recognizes that procedural characteristics do not influence all applicants' in the same way, but rather operate in an integrative fashion with the applicants' personal characteristics.

Participants
Data were collected from 304 applicants (73% female; Mean age of 20.4 years) who were applying for entry into primary (n = 157) or secondary (n = 147) ITT programs in UK universities.The majority of participants selfidentified as white (97%) and without a disability (92%).

Measures and Procedures
We invited prospective teacher trainees applying for entry into ITT programs to participate in this study.Participation involved voluntarily participating in the SJT pilot at the end of their interview session.Prior to completing the SJT, participants completed individual and group interviews, a written task, and a skills test in mathematics.Applicants were provided with the following instructions before they began the interview session: As part of the interview, you will be invited to complete a pilot version of the Teacher Situational Judgment Test.The test will take 60 minutes and will be scheduled at the end of your interview.The information collected in this SJT pilot will be used to help develop and refine the test for use in future selection processes.Your responses will not be used for anything other than the research associated with this pilot process.

SJT. Applicant trainee teachers completed a 35-item Situational Judgment
Test.Each item consisted of one scenario with 5 or 8 outcome options (see Figure 2 for an example SJT scenario).The paper-and-pencil test consisted of one test booklet with two parts and a separate answer sheet.In Part I, participants were presented with 20 scenarios, each with 5 possible response options to rank (from most appropriate to least appropriate).Part II consisted of 15 scenarios, each with 8 options.Participants to select the 3 most appropriate options that, when used in combination, would represent the best course of action.Before participating in the pilot SJT, participants were asked to sign a consent form, which indicated whether they consented to the results of the SJT being linked with feedback from entry interviews and course performance for future SJT validation.Participants were assured  that the pilot SJT results would not be used for current program selection purposes.
You are teaching a lesson and have asked the students to individually complete an exercise that requires them to write down their responses.You have explained the exercise to the students and answered all of the questions that they have asked.As the students begin writing, one student, Ruby, starts to throw paper around and is clearly distracting the students sitting nearby.You know from previous incidents that Ruby often becomes frustrated when she does not understand how to complete activities, and that she often displays her frustration by being disruptive.

Choose the three most appropriate actions to take in this situation (alternatively, Rank the items in the most appropriate order)
 Send Ruby out the class if she continues to be disruptive  Ask Ruby if she understands what the activity requires her to do  etc. (eight total response options)

Figure 2. Example of situational judgment test content
Participant evaluation form.Following the SJT, participants were invited to complete a feedback form that consisted of three content-related items and four items concerning the use of SJTs as a method for selection (see Table 1 for item wording).The seven items measured participants' evaluation of (a) content relevance, (b) content difficulty, (c) content fairness, (d) SJT differentiation, (e) SJT fairness, (f) SJT appropriateness, and (g) SJT measurement.Participants scored each item from 1 = Strongly disagree to 5 = Strongly agree, with the mid-point descriptor at 3 = Neither agree nor disagree.In addition, participants were asked to indicate the length of time it took them to complete the SJTs, and were given the opportunity to provide open-ended comments following prompts (e.g., What aspects did you find most/least effective?and How does it compare to other selection methods you have experienced?).The SJT is able to measure the attributes that are necessary for trainee teachers 3.11 (1.07)

Analyses
Our analysis of the feedback data involved a two-step process: quantitative analysis (means and standard deviations) of the 304 participants' ratings of the seven content and SJT-specific items and content (qualitative) analysis of the written comments from the 118 participants who provided responses to the open-ended items.Broad a priori themes from the content analysis included relevance, fairness, self-perception, test design, and context, while further analysis of the individual comments revealed themes based on organizational justice theory, and comprised two categories: (a) evaluation of the SJT as part of the selection procedure (procedural justice), and (b) evaluation of the fairness of using the SJT to inform decision making (distributive justice).Open-ended comments were further coded with themes of personal characteristics (i.e., work experience, personality).

Results
The majority of participants (90.6%) completed the 35-item SJT within 60 minutes, with a mean completion time of 49.1 minutes (SD = 18.4).Table 1 presents the item means and standard deviations for the three content items and the four method items, revealing scores above the mid-point and ranging from 3.08 for SJT appropriateness to 3.81 for content relevance (on a 5-point scale).There were no significant group differences for gender or for teaching level (primary or secondary) for overall content or method.Primary applicants provided higher ratings across all seven items, with significantly higher means on two method items: fairness F (1,212) = 4.82 (p = .04)and appropriateness F (1, 212) = 5.75 (p = .019).Of the 304 participants, 118 participants (55 primary, 63 secondary) offered both quantitative and qualitative feedback on the SJT.There were no significant rating differences between applicants who offered feedback on the open-ended items and those who did not.
We asked for feedback about the effectiveness of the SJTs as a selection method.Of the participants who provided written feedback (N = 118), 74.6% provided specific comments critical of test design (e.g., too long; scenario options too similar) and context (e.g., scenario options based more on secondary teaching than primary).The remaining participants (25.4%) provided positive and general evaluations (e.g., very thorough; good variety of scenarios; appropriate and effective for selection).
Procedural and distributive justice issues.Of the 118 participants who provided written feedback, 58.4% expressed procedural justice concerns associated with inexperience (e.g., "hard to judge if you've never been in a certain situation before") and social desirability (e.g., "people could answer them the way they feel they are expected to as opposed to their own true reaction to the situation").Participants who lacked confidence in the SJT as a suitable and effective selection method expressed concern because there was "no human interaction" or no opportunity for "the justification of responses."This group of participants questioned the effectiveness of the SJT as a selection tool because the design did not allow "the personality" of an individual applicant to emerge.

IJEP -International Journal of Educational Psychology, 3(2) 117
Content analyses also revealed comments on the indirect nature of the SJT, which is designed to capture implicit traits as well as expressed, explicit traits.For example, an applicant questioned whether the test "truly reflects ability" while another stated, "many answers were based on 'gut feel' rather than knowledge."Participants found the SJT "useful in some respects to the aptitude of teaching" but also expressed distributive justice concerns about the fairness of outcome decisions based on hypothetical situations that involved skills not yet developed (e.g., "unless [you] faced the situation for real you don't know exactly how you would react").There was a fear that "too much [would] be read into the results" since the SJT appeared to be measuring skills that "should be [taught] during teacher training."Practical issues.Practical recommendations offered by participants, centered on test design (e.g., present using a computer instead of paper-andpencil; include opportunities to write open-ended responses), teaching context (e.g., a need to create separate SJTs for primary and secondary program applicants), and selection use (e.g., appropriate assessment when used in combination with other tools; could be useful as a method to focus on specific characteristics of an applicant to question them further in an interview).

Discussion
Previous research has not explored how applicants to teacher training programs perceive the fairness and appropriateness of the selection process.Applicants' perceptions of the selection process should be considered because how applicants evaluate the fairness of their experience influences the long-term success of the selection process (Patterson et al., 2011).Results from this study provide support for the use of SJTs as part of the selection process, particularly if the tools are tailored to primary or secondary contexts.Primary applicants provided higher ratings than secondary applicants overall, yet some comments included concerns that the SJT was biased to those familiar with a secondary school context.Our own previous work shows a general tendency for primary teacher trainees to provide higher overall scores on rating scales than secondary teacher trainees (e.g., Klassen, Perry, & Frenzel, 2012), but the result is puzzling and  deserves further exploration in future studies.Some applicants found it "hard to answer some of the questions when you only know a limited amount of information.It would be easier to work out the right thing to do if you were there and knew the context."Although school contexts are multidimensional, with a wide range of school types, teaching subjects, socioeconomic factors, and urban-rural settings, our findings suggest that teaching level is particularly important dimension for prospective teachers.In an organizational justice context, procedural justice is well-served if SJTs are tailored to the teaching level contexts of applicants.Patterson et al. (2011) found that applicant reactions to various selection methods (an SJT, a knowledge test and a selection centre) were most positive for the higher-fidelity methods.Some of our applicants stated a preference for higher-fidelity methods (such as teaching simulations) over lower fidelity methods (such as SJTs), with concerns of not being fairly evaluated because they perceived SJT scenarios as based on classroom experience (something most did not have yet).But since one goal of the SJT is to measure tacit and procedural knowledge (Elliott et al., 2011) in a specific domain (i.e., interpersonal skills), it is expected that ITT applicants complete the test without access to teaching-specific knowledge.Specific teaching experience is therefore unnecessary at the time of application for the SJT to assist predicting future teacher-student practicum interactions and job performance (Lievens & Sackett, 2012).
Some secondary participants expressed concerns about response distortion, or 'faking.'Two factors make SJTs less susceptible to faking than conventional personality tests.First, SJTs can be constructed to be less susceptible to faking by using cognitively loaded formats that present candidates with multiple domains with heterogeneous content (Patterson, Ashworth, Kerrin, & O'Neill, 2013).Second, on our SJT, applicants were asked what they should do in the given situation, rather than what they would do, resulting in lower susceptibility to faking.SJTs measure implicit trait policies that reflect beliefs about the costs and benefits of expressing particular traits in certain situations.In other words, applicants choose what they believe is the best response-not their behavioural tendency-meaning that faking is less relevant.Compared to conventional personality tests, SJTs are less prone to faking, and deliberate faking only negligibly changes scores (Lievens, Peeters, & Schollaert, 2008).Future research on the relationship of implicit trait policies and actual teaching behaviours is needed.In addition, (as highlighted by Patterson, Ashworth et al., 2013), further work is needed to explore how the interaction between cognition, personality, and emotions is related to training outcomes and job performance.
Practical implications.Assessing implicit traits through indirect measurement is not without criticism and, like Haines and Sumner (2013), we view critiques as one way to improve measures.By piloting the SJT and providing an opportunity for applicants to provide feedback, we were able to apply critiques towards the use of implicit measures in real world application settings.The participants in this study offered several practical recommendations for refining the SJT for future ITT candidate selection.One suggestion-to develop context-specific (primary, secondary) SJTs-is already being implemented.Although we do not have plans for shortening the SJT, we believe that the use of alternate testing means (video-based; computer screen) will help alleviate some of the test fatigue expressed by participants.For example, Patterson et al.'s (2012) review found video-based SJT formats generally received equal or more positive ratings when compared to written paper-and-pencil formats.Our results included applicant recommendations for different formats (i.e., on a screen, use of computer, without separate answer sheet, open-ended responses).
Since participants expressed fairness concerns based on a perceived disadvantage due to lack of teaching experience, future SJTs for ITT entry can include a statement that teaching knowledge is not needed to complete the SJT.This statement is supported by the literature (e.g., Lievens & Sackett, 2012), which identifies the SJTs as a measure of implicit procedural knowledge like interpersonal skills.Being notified of how the SJT results will be used during the selection process (e.g., in combination with an interview and other selection activities) may also help alleviate procedural concerns.The advantages of using SJTs as part of the selection process into teacher training are many: increased predictive and incremental validity, low cost, reliability, and a proven track record in the selection process in other disciplines.There is clearly much more work needed to improve the teacher training selection process, but continued attention to issues of procedural and distributive justice is warranted to ensure that the selection process is not only fair to all applicants, but perceived to be fair by applicants.

Figure 1 .
Figure 1.An integrated model in which situational judgment tests influence applicants' personal characteristics and perceived procedural characteristics.Note.The model was influenced by previous work by Gilliland (1993), Hausknecht et al. (2004), and Patterson et al. (2011).
in building a 35-item pilot SJT for Initial Teacher Training (ITT) selection.The SJT scenarios are contextualized-set in a school setting-and were developed through discussions with expert teachers, usually ITT mentors working closely with trainee teachers, utilizing critical incident technique.We followed a series of development steps in 2013 by working closely with teacher subject matter experts (SMEs), school authorities, and university-based teacher education directors.The development steps represent a practice-based approach to domain and content development, resulting in three broad non-cognitive domains:  Empathy and Communication -Candidate demonstrates active listening, and engages in an open dialogue with both pupils and colleagues.Candidate seeks advice pro-actively and is responsive to both professional feedback and pupil's needs.Candidate has the ability to adapt the style of communication and nature of dialogue appropriately. Resilience and Adaptability -Candidate demonstrates the capability to remain resilient under pressure.Demonstrates adaptability, and an ability to change lessons and the sequence of lessons accordingly where required.Has an awareness of their level of competence and

Table 1
Participant Evaluation Responses for Teacher Situational Judgment Test