Teachers' Effect on Students' Creative Self-Beliefs Is

Learning and Individual Differences 44 (2015) 1–8

Contents lists available at ScienceDirect

Learning and Individual Differences journal homepage: www.elsevier.com/locate/lindif

Teachers' Effect on Students' Creative Self-Beliefs Is Moderated by Students' Gender Maciej Karwowski ⁎, Jacek Gralewski, Grzegorz Szumski Department of Educational Sciences, The Maria Grzegorzewska University, 40 Szczesliwicka St., 02353 Warsaw, Poland

a r t i c l e

i n f o

Article history: Received 18 May 2015 Received in revised form 13 September 2015 Accepted 4 October 2015 Available online xxxx Keywords: Creative self-beliefs Teachers' influence Teachers' ratings Gender effects

a b s t r a c t Creative self-beliefs, such as creative self-efficacy, predict creative activity and achievement. Still little is known, however, about the factors that shape such self-beliefs. Drawing on Bandura's sociocognitive theory, this longitudinal study tests the role of teachers' expectations on students' domain-specific creative self-efficacy. Teachers' ratings of students' creativity were substantially related to students' creative self-perception a semester later and this effect was significantly stronger among female than male students. We discuss these findings in terms of the accuracy of teachers' beliefs and the consequences of their influence on students' creative self-perception. © 2015 Elsevier Inc. All rights reserved.

1. Introduction The sense of self-efficacy (Bandura, 1997) is one of the drives that initiate and sustain individual activity (Cervone & Peake, 1986). Consequently, it helps to deal with various tasks, including those that require creative thinking. Creative self-efficacy (CSE) is an individual's conviction that he or she is able to manage in situations that require creativity (Beghetto, 2006; Tierney & Farmer, 2002, 2004). CSE not only positively relates to creative personality (Karwowski, 2012; Silvia, Nusbaum, Berg, Martin, & O'Connor, 2009), innovative behavior (Hsu, Hou, & Fan, 2011), and creative achievement (Batey & Furnham, 2008; Tierney & Farmer, 2004), but also mediates the relations between potential and creative achievement (Lim & Choi, 2009). A recent review (Karwowski & Lebuda, in press-a) and a meta-analysis (Karwowski & Lebuda, in press-b) have demonstrated that although creative self-beliefs are related to personality – mainly openness – they are discriminatively and incrementally valid and not reducible to personality traits. CSE shapes itself under the influence of a wide range of factors – psychological, i.e. personality-related (Karwowski, Lebuda, Wisniewska, & Gralewski, 2013), as well as social, i.e. in-class comparisons (Karwowski, 2015a). CSE crystallizes in approximately 10-year-old children (Karwowski, 2015b): Younger children usually tend to associate creativity more with activity and products than with individual traits (Karwowski & Barbot, in press). Although parents and teachers may exert special influence on the shaping of children's and young people's CSE, the creativity literature has not really investigated this yet. In this paper, we intend to fill this gap. More specifically, this ⁎ Corresponding author. E-mail address: [email protected] (M. Karwowski).

http://dx.doi.org/10.1016/j.lindif.2015.10.001 1041-6080/© 2015 Elsevier Inc. All rights reserved.

paper explores the relationship between teachers' ratings of students' creativity and students' creative self-efficacy. Consistently with sociocognitive theory (Bandura, 1997), we perceive teachers as one of the main sources that shape students' creative self-perception and we theorize that not only teachers' perceptions could shape students' creative self-efficacy, but also that this effect will be stronger among female than male students. This general hypothesis drives a study presented below. We start by shortly reviewing the relevant literature about gender differences in creative self-beliefs. In particular, we look for possible causes of the so-called “male-hubrisfemale-humility bias” (Furnham, Hosoe, & Tang, 2002) and show that males tend to overestimate their abilities, while females underestimate them. We theorize that this bias may be (at least partially) caused by differences in teachers' expectations regarding male and female students' creativity, which, consequently, stem from gender stereotypes (Baer & Kaufman, 2008) and teachers' implicit theories of creativity (Gralewski & Karwowski, 2013; in press). We briefly review the wide and diverse literature that suggests the existence of more space for male than female students' creativity at school and argues that teachers' implicit theories of creativity are masculinized. Keeping in mind the higher level of adaptativeness and conformity in female students (von Wittich & Antonakis, 2011), we also expect that female students will be more sensitive to teachers' perceptions and expectations – a hypothesis coherent with previous studies (Baer, 1997; Butz & Usher, 2015; Correll, 2001). Despite the lack of systematic intergender differences in the average creative potential (Baer & Kaufman, 2008; Harris, 2004, but see Karwowski et al., in press, for a discussion about intergender differences in variability of creative potential), males' advantage over females in real-word creative accomplishment is well-established (Abra & Valentine-French, 1991; Helson, 1990; Piirto, 2004), similarly to more

2

M. Karwowski et al. / Learning and Individual Differences 44 (2015) 1–8

positive creative self-beliefs among men, which came to be known as the male hubris-female humility bias (Furnham et al., 2002; Karwowski, 2011). Domain-specific analyses show that men are inclined to assess their creative abilities higher than women in science-analytic creativity (Kaufman, 2006), problem solving (Hughes, Furnham, & Batey, 2013), and creativity in sports (Hughes et al., 2013; Kaufman, 2006), whereas women assess their creative abilities higher in the social area (Kaufman, 2006) and the arts (Hughes et al., 2013; Kaufman, 2006). Even though previous studies have not revealed teachers' influence on students' creative self-beliefs, such a relation is a natural consequence of the expectation models (Darley & Fazio, 1980; Jussim, in press; Urhahne, 2015) and appears to follow from out-of-school studies as well (Tierney & Farmer, 2004). Tierney and Farmer (2004) showed that supervisors' expectations of employees' creativity shape the latter's creative self-efficacy, and are predictors of supervisors' supportive behaviors toward their employees. 1.1. Teachers' expectations, practices, and implicit theories of creativity Although creativity scholars usually agree on the definition of creativity and perceive it as a human capacity to develop ideas that are both novel and appropriate (Amabile, 1996; Sternberg & Lubart, 1999), complexity of creativity makes it a tough subject of scientific studies. Several taxonomies of creativity focusing on different aspects of creativity have been proposed over the decades, (see Glăveanu, 2010, 2014, 2015 for a discussion). From the perspective of this paper, apparently the crucial distinction is that between creative potential (Karwowski, 2015c) and creative achievement (Carson, Peterson, & Higgins, 2005). Asking teachers about how they perceive creativity of their students is obviously asking a question about their creative potential rather than their creative achievement. Such potential, understood as the conglomerate of cognitive and personality characteristics: imagination, divergent thinking, openness, is closer to the mini-c or little-c creativity (Karwowski, 2009; Kaufman & Beghetto, 2009) than creativity visible in observable products. Teachers' expectations with regard to the creativity of their students at least to some extent are a natural consequence of teachers' implicit theories of creativity. Previous studies showed that frequently, teachers quite selectively define creativity and creative students (Andiliou & Murphy, 2010; Gralewski & Karwowski, in press; Karwowski, 2007; Westby & Dawson, 1995), and that their implicit theories differ to a large extent from explicit theories formulated by experts (Dawson, D'Andrea, Affinito, & Westby, 1999). Keeping in mind that creativity researchers also quite often disagree on the definition of creativity (Plucker, Beghetto, & Dow, 2004; Kaufman, in press), teachers' lack of coherence itself is neither surprising, nor especially damaging. However, the consequences of those discrepancies may be longstanding. A recent study has demonstrated (Gralewski & Karwowski, in press) that about one-third of participating teachers defined creativity oppositely to existing theories of creativity (e.g., perceiving creative students as conformist, subordinated, having not too original ideas), while the remaining two-thirds perceived creativity either in terms of revolutionary (Gilson & Madjar, 2011) and innovative (Kirton, 1976) creativity (dominating pattern – 46%) or in terms of more incremental (Gilson & Madjar, 2011), adaptive creativity (20%). What is even more interesting, this study also showed that those teachers who defined creativity in terms of innovativeness better recognized creative abilities of male students, while those who perceived creativity as adaptive focused more on female students' creativity and better recognized their abilities. Previous studies have also shown that teachers more often accept boys' individualism and independence while expecting girls to be rather persevering, subordinate, cooperative, and able to take care of interpersonal relationships (Bianco, Harris, Garrison-Wade, & Leech, 2011). Boys are also more likely to have the right to show weaker control over their impulsiveness, whereas girls are expected to present a more

mature approach that manifests itself in the ability to create and sustain relationships with other people. Teachers reward girls for their ability to cooperate, complete tasks, and behave appropriately in the classroom (Lindley & Keithley, 1991). Therefore, in school settings, boys may likely have more opportunities to develop independence and autonomous thinking, whereas girls undergo a peculiar training in perseverance and subordination to teachers (Forgasz & Leder, 2001; Robinson & Lubienski, 2011). These differing expectations may also shape students' diverse convictions about their own skills and creative abilities. Beghetto (2006) demonstrated that teachers' feedback on creativity was the strongest predictor of students' CSE, but the author did not analyze the possible gender-specificity of this effect. Regardless of the possible causes of differences in self-perception among males and females, there are at least two additional reasons to believe that female students are influenced by teachers' expectations to a greater extent than male students are (Correll, 2001). First, due to their higher school engagement and the greater value they attach to the relations with their teachers (Rudasil, Niehaus, Buhs, & White, 2013; Wolter, Gluer, & Hannover, 2014), female students are more responsive to teachers' messages and behaviors. This likely involves creativity as well. Previous research (Baer, 1997) has demonstrated that while engaging into creative tasks girls are more sensitive to teachers' evaluation than boys. Second, previous studies have found a small, yet systematic difference between male and female students in their creative mindsets (Karwowski, 2014) – the perception of creativity as malleable versus fixed. As male students tend to perceive creativity as a more fixed and less malleable characteristic than female students do, their sensitivity to external influences on their CSE may be generally lower. As in their perception creativity is generally stable, the possibility of teachers' influence is lower among them. 2. The present study This study aims at filling the gap in the research on relationship between social conditions and students' creative self-beliefs. Two specific hypotheses drive the study: H1: Teachers' expectations about students' creativity positively predict students' creative self-beliefs and H2: Students' gender moderates the strength of the relationship between teachers' expectations about students' creativity and students' creative self-beliefs. More specifically, we hypothesize this relationship to be stronger among female than male students. 2.1. Method 2.1.1. Participants One group of participants was made up of 1614 first-grade middleschool students (Mage = 13.15, SD = 0.44, 49% females), members of 80 classes, attending 40 schools across Poland. The other group of participants were their teachers (N = 189). Two of them rated each student: A Polish language teacher (n = 97, 51%) and a math teacher (n = 92, 49%). As the study was longitudinal (see the Procedure section), there were some drop-outs between waves and the number of missing values differed across variables. We therefore chose to impute missing data using the multiple imputation option available in Mplus software. There were 172 female teachers (94%) and 12 male teachers (6%), 5 teachers did not answer the question about gender. Teachers' were between 25 and 64 years old, with M = 44.76 and SD = 8.50. 2.1.2. Instruments 2.1.2.1. Teachers' ratings of students' creativity. Teachers described all students in the class in terms of a wide variety of characteristics, including two items developed specifically for the purpose of measuring perceived creativity of students. The first variable was the level of creativity estimated on an IQ-type scale (M = 100, SD = 15). Teachers were provided with a short explanation of normal curve characteristics


and asked to assess the creativity of each student in the class by using this scale (see Furnham et al., 2002; Kaufman, 2012, for a similar approach). The second question asked teachers to assess whether each of the students in the class “had a lot of ideas,” using a 5-point Likert scale, with 1 = not at all, and 5 = definitely yes. Each student was rated by two teachers during Time 1; teachers' ratings were modeled as a latent variable. 2.1.2.2. Creative self-efficacy in math and language. A new scale to measure domain-specific creative self-efficacy in the domains of math and Polish language was created for the purposes of this study. Ten items were developed, five devoted to creative self-efficacy in Polish language and five to creative self-efficacy in math, with a 0–100 percentage scale and an instruction for the participants to assess how strongly they were convinced they would be able to perform the activities listed (see Bandura, 2006, for an overview and recommendations regarding such scales). All the items and psychometric properties of this scale are presented in Table 1. 2.1.2.3. Creative potential. Creative potential was measured by five divergent thinking tasks (Guilford, 1967): two “unusual uses” tasks (generating unusual uses for toothpaste and ice-cream), two consequences tasks (what would happen if there were no cell phones or Internet and what would happen if people ate only candy), and one similarities task (what is similar about a bar and a computer). Each task was scored for fluency, and so was the number of answers provided. Fluency reliability index was very good (α = .86). The creative potential was measured during Time 1 (September). For the purpose of analyses it was modeled as a latent variable composed of five observed fluency scores. 2.1.3. Procedure The study was longitudinal, with two measurement waves – the first one (Time 1) at the beginning of the school year (September) and the second one (Time 2) at the beginning of the second semester (February). Teachers' ratings and students' creative potential were measured during the first wave, together with several teachers' and students' characteristics that are outside the scope of this study. Students' creative self-efficacy was measured in the second wave. Students filled their questionnaires in class, in a group setting. Informed consent was obtained from all the teachers and students as well as from school principals and students' parents. 2.1.4. Results The results are analyzed in four steps. We start with providing some details about the psychometric characteristics of self-efficacy measurement, then we present basic descriptive statistics and intercorrelations. In the third step, we provide details on gender differences in the variables used in this study and then we test hypothesized gender moderation effects, in the fourth step we switch to our main research questions and present structural equation models testing for gendermoderated effects of teachers' ratings on students' creative self-efficacy.

3

2.1.4.1. Creative Self-Efficacy in Math and Language. To explore the structure of creative self-efficacy measurement, we followed a twostep procedure. First, we divided the total sample into two random subsamples. We put the first subsample to the exploratory factor analysis with oblimin rotation. Two factors were extracted on the basis on the scree plot and eigenvalues N 1, and together they explained 65% of total variance (see Table 1). As may be observed, there were clear math and language CSE factors. Confirmatory factor analysis performed on the second subsample confirmed the two-factor structure (CFI = .942, RMSEA = .069, 90% CI: .063, .076). Additional analyses were performed within both scales (self-efficacy in language and self-efficacy in math) to confirm their unidimensionality. It was indeed confirmed in the case of creative self-efficacy in language (CFI = .995, RMSEA = .047, 90% CI: .037, .058) and creative self-efficacy in math (CFI = .969, RMSEA = .080, 90% CI: .069, .091). The latent correlation between the two aspects of self-efficacy was robust and significant r = .60, yet still in favor of their discriminant validity (Brown, 2015). 2.1.4.2. Descriptive Statistics and Intercorrelations. Descriptive statistics and intercorrelations between all the variables used in this study (with creative self-efficacy scores calculated as observed factors: α = .90 for creative self-efficacy in math and α = .87 for creative selfefficacy in language) are presented in Table 2. Both within- and between-teacher correlations of student ratings were significant and moderate (between-teacher) to high (withinteacher). Teachers' ratings were significantly yet weakly correlated with students' creative potential. Even more importantly, teachers' ratings correlated significantly with students' creative self-efficacy in math and language. 2.1.4.3. Gender Differences. As our hypothesis assumed gendermoderated effects of teachers' ratings on students' creative selfefficacy, we began with examining whether male and female students differed in creative self-efficacy in math and language. Males obtained significantly higher scores than females in math CSE (M = 56.6, SD = 23.70 and M = 49.47, SD = 23.36, respectively; F[1,1328] = 30.48; p b .001), while females perceived themselves as more creative than males in the domain of language (M = 65.30, SD = 21.49 and M = 60.94, SD = 21.18, respectively; F[1,1328] = 13.85; p b .001). Although among males and females self-perceived creative self-efficacy was a significantly higher in language than in math (females: t[661] = −18.55, p b .001, males: t[664] = − 5.61, p b .001), the Gender x CSE type interaction was significant (F[1,1325] = 97.83; p b .001, η2 = .069; see Fig. 1). As teachers' ratings were estimated in a complex way, we did not compare gender differences in each of the observed variables, but instead, in the first step we fitted a one-factor CFA model with teachers' ratings, and in the second step we examined the factor invariance of this model across students' gender. The model fit was almost perfect (CFI = .994, TLI = 1.000, RMSEA = .000, SRMR = .009) and the model

Table 1 Creative Self-Efficacy in Polish Language and Math – Items, Descriptive Statistics, and Factor Loadings. Using a 0-100 scale, please rate how convinced are you that you could effectively deal with:

M

SD

Skewness

Kurtosis

CSE-math

CSE-lang.

...solving a math problem in an original way – different from what the teacher proposes. ...writing an inventive, interesting paper for a Polish class. ...coming up with at least a few different ways of solving a word problem in math. ...imagining a situation described in a math problem. ...imagining a situation described in a poem discussed during a Polish class. ...coming up with your own interesting math problem. ...discovering the rule that will make it possible to solve a math problem that is outside of your math curriculum. ...coming up with an ending for a story you began reading in a Polish class. ...writing a short story. ...being inventive during math classes. ...being inventive during Polish classes. ...discovering by yourself a grammar rule that will make it possible to remember how to write difficult words.

53.06 62.49 50.44 61.13 64.87 54.11 47.03 65.98 69.86 55.12 61.79 54.09

28.09 26.46 27.89 28.10 26.90 30.15 30.07 27.78 27.75 29.97 26.96 28.86

-0.05 -0.41 -0.02 -0.36 -0.52 -0.14 0.06 -0.55 -0.75 -0.15 -0.43 -0.15

-0.76 -0.59 -0.80 -0.70 -0.50 -1.00 -0.98 -0.57 -0.29 -1.01 -0.58 -0.84

0.83 0.07 0.88 0.59 0.11 0.77 0.90 -0.10 -0.16 0.78 0.14 0.27

-0.04 0.78 -0.03 0.29 0.72 0.07 -0.12 0.88 0.90 0.10 0.74 0.48

4


Table 2 Descriptive Statistics and Intercorrelations Between Variables. Variable 1 2 3 4 5 6 7

Teacher 1 - many ideas Teacher 1 – creativity Teacher 2 - many ideas Teacher 2 – creativity CSE - Math CSE - Language Fluency

Min

Max

M

SD

2

3

4

5

6

7

3.1

1.1

.64

.36

.35

.21

.22

.17

130 5

98.3 3.0

19.3 1.1

.36 1614

.37 .61

.23 .24

.21 .23

.16 .11

160 100 100 111

98.4 53.5 63.2 15.28

19.5 23.7 21.4 9.90

1049 1175 1175 1318

.19

.22 .54

.15 .12 .15

1

5

70 1 50 0 0 0

1200 1121 1121 1096

722 722 1250

1441 813

1347

All the correlations are significant at the level of p b .001 or higher. Consecutive sample sizes for correlations are presented below the diagonal.

was invariant across genders (configural invariance model: CFI = .999, TLI = .991, RMSEA = .041; metric invariance model: CFI = .991, TLI = .979, RMSEA = .064, scalar invariance model: CFI = .995, TLI = .993, RMSEA = .038). The difference between configural and metric model fit was not significant (Δχ2(Δdf = 3) = 7.37, p = .06), and neither was the difference between scalar and configural models (Δχ2(Δdf = 6) = 8.08, p = .23) or between scalar and metric models (Δχ2(Δdf = 3) = 0.709, p = .87). Thus, not only the structure of creativity perceived by teachers was the same across genders (configural invariance), but factor loadings of items used (metric invariance) and item intercepts (scalar invariance) were also similar. After establishing this, we checked whether teachers' ratings were associated with students' gender. The model with gender predicting the latent variable of teachers' ratings was almost perfectly fitted (CFI = 1.000, TLI = .999, RMSEA = .009, SRMR = .009), with gender (coded: 0 = females, 1 = males) being significantly associated with teachers' ratings (β = −.103, SE = 0.049, p = .036 – an equivalent of Cohen's d = − 0.207). Hence, although the difference between teacher-rated creativity of female and male students was small, it was significant and showed that, overall, teachers perceived females as more creative than males. A comparison of fluency scores showed that female students outperformed their male classmates: M = 15.16, SD = 10.95 and M = 13.59, SD = 8.35 respectively, F[1,1596] = 53.97, p b .001, yet the effect size of this difference was small (Hedges g = 0.37, 95% CI: 0.27,0.47).

2.1.4.4. Structural Equation Models. Our final and main analyses dealt with the question of whether and to what extent teachers' rating of students' creativity predict students' creative self-efficacy in language and math after a semester, and while controlling for students' creative potential, and whether this effect was moderated by students' gender (Fig. 2).

The model fitted the data well: χ2(df = 360) = 1401.96; χ2/df = 3.89; CFI = .929, RMSEA = .042, 90% CI .040, .045. In all cases, the observed variables were strongly loaded by latent constructs, which confirms their construct validity. Additionally, the composite reliabilities were similar across male and female students (Table 3). In the case of both self-efficacy measures, the regression coefficient between teachers' ratings and female students' CSE was about two times higher than between teachers' ratings and male students' CSE. To test the moderating effect of gender, we compared the effect sizes (both standardized and unstandardized) as well as confidence intervals around them among males and females. In the case of the relationship between teachers' ratings and students' creative self-efficacy in language the standardized effect size was estimated at β = 0.58 for females and β = 0.18 for males. The unstandardized coefficient for females was estimated at B = 22.18 with SE = 3.19, which gives a 95% confidence interval between 15.93 and 28.43. For males, the unstandardized coefficient was B = 5.56, with SE = 1.91 and 95% CI: 1.82–9.30. As these intervals do not overlap, there was a clear difference in the slopes (z = 4.47) and, consequently, a significant moderation effect. The analysis performed for creative self-efficacy in math yielded similar results. The standardized effect size in math was estimated at β = 0.49 for females and β = 0.32 for males. The unstandardized coefficient was estimated at B = 19.60 for females (SE = 3.12, 95% CI: 13.49–25.72) and at B = 11.82 for males (SE = 2.42, 95% CI: 7.08–16.56). Although these confidence intervals slightly overlapped, also this difference is statistically significant (z = 1.97), and shows a gender-moderated effect. Students' creative potential measured by divergent thinking tasks was significantly related to teachers' ratings at Time 1, being additionally stronger associated with perceived creativity of female than male students. Interestingly however, such operationalized potential did not predict students' own self-perception after a semester, except a weak effect observed in case of males CSE in the sphere of language.

Fig. 1. Gender differences in creative self-efficacy in the Polish language and math.


5

Fig. 2. The longitudinal relationship between teachers' ratings of students' creativity and students' creative self-efficacy in math and the Polish language. F: females, M: males.

3. Discussion Creative self-efficacy is a powerful source of curiosity and creative interests (Karwowski, 2012), which motivates people for creative activity (Tierney & Farmer, 2002), and hence increases their chances for creative achievement (Tierney & Farmer, 2011). Although there is a growing number of empirical studies devoted to domain-general creative self-efficacy (Beghetto, 2006; Karwowski, 2011) and domainspecific self-perceptions (Kaufman, 2006; Pretz & McCollum, 2014), creative self-efficacy in school domains has been only scarcely studied (Beghetto, Kaufman, & Baxter, 2011). However, the focus on schoolrelated creative self-efficacies seems to be especially important when one keeps in mind the problems creative students face at school (Beghetto, 2014). In the study presented in this article, we were especially interested in the relationship between teachers' perception and students' creative self-efficacies. What bothered us even more was the extent to which this relationship may be moderated by students' gender. The sociocognitive theory (Bandura, 1997; Karwowski & Barbot, in press) locates the factors that shape people's self-efficacy mainly in the social environment. Besides mastery experiences and physiological experiences, the strongest factors that shape self-efficacy are the so-called vicarious experiences and social persuasion. A clear example of such experiences lies in teachers' expectations and behavior toward students. Therefore, we theorized that teachers' perception of an individual student will be related to her or his creative self-efficacy. The longitudinal study demonstrated the hypothesized moderating effect of students' gender on the relationship between teachers' ratings and students' creative self-efficacy, even when we controlled the initial level of creative potential. As we hypothesized, this relationship was significantly stronger for female than male students. We also observed Table 3 Construct Reliabilities of Latent Variables in SEM Model.

Females Males

Teachers’ Nomination H

Students Creative self-efficacy Math H

Students Creative self-efficacy Lang H

Fluency

.69 .70

.91 .90

.90 .86

.91 .86

that males perceived themselves as more creative in mathematics, while females perceived themselves as more creative in the sphere of language. This pattern fits well into the wide body of research that reveals differences in domain-specific self-concepts (Marsh & Yeung, 1998; Gentile et al., 2009). Thus, it may be argued that the pattern of differences in school-domain-specific creative self-concepts is the same as in the case of academic self-concepts. This is not to say, however, that creative self-efficacy in the domain of math or language is the same as an academic self-concept in these domains. A subsequent study (Karwowski & Szumski, 2015) shows that although the correlations between creative and academic self-efficacies in school subjects are substantial (about r = .60), these constructs are still distinguishable and their measurement is characterized by satisfactory discriminant validity. While previous research and theories (Gralewski & Karwowski, 2013; in press) provide a good reason to believe not only that males perceive themselves as more creative than females, but also that teachers perceive male students as more creative and gifted than female students (Li, 1999), this study yields the opposite result. Contrary to our expectations, when we asked teachers to rate the creativity of their students in school subjects, they perceived female students as significantly more creative than male students. This finding suggests that the perceived male superiority in creativity is not universal and applies mainly to out-of-school domains of creativity. Even more interestingly, the correlation between teachers' ratings and students' creative potential observed in this study was stronger for female than male students. Although this finding is in line with previous research that shows that teachers' are better at recognizing females' than males' creativity (Sommer, Fink, & Neubauer, 2008), it stands in opposition to previous studies conducted in Poland (Gralewski & Karwowski, 2013; in press). Finally, our study provides convincing arguments that female domain-specific creative self-efficacies are much more strongly related to their teachers' perception than is observed among male students. Therefore, our hypothesis that students' sensitivity to teachers' expectation is gender-dependent was supported. It should be kept in mind, however, that in this medium-term longitudinal study the effect of teachers' expectation was significant among both male and female students, which showed that, indeed, social persuasion and vicarious experiences (Bandura, 1997) are powerful factors that shapes students'

6


self-view. Even among males this effect was quite substantial (especially in the case of creative self-efficacy in math), while among females both effects were strong according to the usually applied criteria (Cohen, 1992). Therefore, not only was females' self-perception more teacher-dependent than males' self-perception, but this effect was also sufficiently strong to potentially play a role in real everyday school life. A stronger relationship between teachers' expectations and females' and males' self-perceptions requires further studies and convincing theoretical explanations. Previous studies in educational psychology demonstrated that parents' (Simpkins, Fredricks, & Eccles, 2012) and teachers' effects on students' self-concept are robust, although usually no gender-moderated effects were observed. The exceptions worth mentioning include the study that showed that the father's gender stereotype perception influenced the child's math interests and that this effect was stronger (and negative) for girls and weaker (and positive) for boys (Jacobs, Davis-Kean, Bleeker, Eccles, & Malanchuk, 2005). It may be speculated that boys' higher “immunity” to teachers' or parents' perception stems from their higher independence and from the lower value they attach to fulfilling the role of a “good student” (Gabriel & Gardner, 1999). It cannot be ruled-out, either, that this effect is at least in part caused by the different creative mindsets boys and girls have (Karwowski, 2014). Previous studies demonstrated that although in childhood students tend to believe in the malleable character of their abilities and believe they are able to develop their skills, their fixed mindset increases with age (Freedman-Doan et al., 2000). Studies in the creative mindset demonstrated that males tend to hold higher fixed-creative and lower growth-creative mindset than females do (Karwowski, 2014). Thus, as they perceive their creativity as hardly changeable and generally stable, the weaker impact of social factors on the level of their creative self-concept is not surprising. However, both these hypotheses require future studies. Finally, this study focused more on domain-specific (math and language) creative self-efficacies, and suggests that the role of teachers' expectations may be especially strong in the case of school- and subjectbased teachers' perceptions and creative self-efficacies. On the one hand, this finding supports arguments that it is much more empirically fruitful to focus on domain-specific than on domain-general creative self-efficacy (Baer, in press). Further, it supports the earlier suggestions (Baer, in press; Meisels, Bickel, Nicholson, Xue, & Atkins-Burnett, 2001) that it makes much more sense to ask teachers about as concrete and observable behaviors of their students as possible instead of using broad categories such as “general creativity.” Last but not least, the results of our study may not necessarily be read as showing the influences of teachers' perceptions on students' creative self-beliefs (causal relationship postulated here and tested in the longitudinal study), but rather in light of the discussion on teachers' ability to recognize abilities of their students (e.g., Baudson, Fischbach, & Preckel, 2014; Ready & Wright, 2011).1 Although over the decades, teachers' ratings (like other informants' ratings, e.g., Renk & Phares, 2004) were used as an independent source of knowledge about children's abilities (Begeny, Eckert, Montarello, & Storie, 2008), in the specific case of creativity those ratings have a rather poor reputation (Baer & Kaufman, 2008; Gralewski & Karwowski, 2013; Karwowski, 2007). Teachers are very effective at recognizing intelligence of their students (Südkamp, Kaiser, & Möller, 2012; Baudson et al., 2014), but – at least in the previous studies – they were not effective at all in recognizing their students' creativity (Brandau et al., 2007; Hoff & Carlsson, 2011; Rudowicz, 2004). If the results of this study are perceived in this light - i.e., correlations between teachers' perception of students' creativity, students' divergent thinking skills, and students' creative self-beliefs are treated as a test of teachers' accuracy in recognizing students' potential, these findings look differently. Overall, teachers' ratings are strongly correlated with students' self-perception, and substantially with results of students'

1

We are grateful to an anonymous reviewer for the suggestion of this line of reasoning.

divergent thinking tests. Moreover, only teachers' ratings were able to predict students' self-perception, while students' abilities were unable to do that (see Karwowski, 2011 and Pretz & McCollum, 2014 for similar findings). Hence, contrary to the previous findings, this study shows that teachers' ratings of creativity may be more accurate than students' perception of themselves. However, even if our findings are read as a voice in the discussion about teachers' accuracy in recognizing students' creativity, significantly higher accuracy in case of female than male students is still open to satisfactory explanation. The differences in regression slopes we have observed – especially large in case of creative self-efficacy in language – are not attributable to the differences in creative potential, as males and females differed in their fluency only slightly. It is possible that those effects are caused by higher achievement of females in language classes. However, although differences in school achievement may stand for perceived differences in average creativity ascribed by teachers to male and female students, those differences are unable to explain the stronger correlations among females. Hence, future research should not only replicate this gender-specific effect, but also propose and test a more robust theory, which may explain this pattern of findings. 3.1. Limitations and future directions The findings presented in this article should be read in the light of their limitations. The first and main limitation lies in the limited sample of teachers who rated students' creativity and the vast majority of female teachers in the sample. Although this proportion resembles the real gender structure in Polish schools (Lebuda, 2014), it may form a confound variable, which influences the observed findings. Hence, future studies should try to replicate these findings by using larger and more gender balanced samples of teachers. Another potential limitation may be our focus on young people: middle-school students. From the teacher expectancy effect studies (Jussim, in press; Kuklinski & Weinstein, 2001; Raudenbush, 1984) we know that these effects are the strongest among students in the beginning of their education – and this applies to the influence of teachers' expectation on students' achievement (Tierney & Farmer, 2002) as well as to their selfperception (Rubie-Davies, 2006). Therefore, such a study devoted to creative self-efficacy in school subjects at the beginning of schooling would be especially important for capturing the developmental aspects of creative self-beliefs (Karwowski, 2015b; Karwowski & Barbot, in press) and the role that social persuasion and vicarious experiences play in their development. 3.2. Conclusions and implications for practice Teachers' perception of students' creativity matters for students' self-perceived creativity and this effect is stronger in the case of female than male students. Therefore, teachers should be especially careful while sending – often unconscious – messages about the perceived creativity of female students, especially if this message conveys the perceived lack of creativity. Otherwise, there is a serious risk that female students will internalize these expectations and that their creative selfefficacy will decrease. The relationship between female students' creativity as perceived by teachers and students' creative self-efficacy is sufficiently strong to consider its role seriously. Obviously, as both classic (Babad, Inbar, & Rosenthal, 1982) and more contemporary works have demonstrated (Jussim, in press; Jussim & Harber, 2005), the effect of expectations works as a double-edged sword and can also bring positive consequences (Hinnant, O'Brien, & Ghazarian, 2009). By strengthening students' creative self-beliefs, teachers simultaneously increase the probability that these students will engage in creative activity. As a consequence, they may achieve creative successes – at school and beyond.


Acknowledgements The research presented in this article was made possible thanks to grants from the National Science Centre, Poland (UMO-2012/07/B/ HS6/01434 and UMO-2012/07/B/HS6/01442). References Abra, J., & Valentine-French, S. (1991). Gender differences in creative achievement: a survey of explanations. Genetic, Social, and General Psychology Monographs, 117, 235–284. Amabile, T.M. (1996). Creativity in context. Boulder, CO: Westview Press. Andiliou, A., & Murphy, P.K. (2010). Examining variations among researchers' and teachers' conceptualizations of creativity: a review and synthesis of contemporary research. Educational Research Review, 5, 201–219. Babad, E.Y., Inbar, J., & Rosenthal, R. (1982). Pygmalion, Galatea, and the golem: investigations of biased and unbiased teachers. Journal of Educational Psychology, 74, 459–474. Baer, J. (1997). Gender differences in the effects of anticipated evaluation on creativity. Creativity Research Journal, 11, 173–177. Baer, J. (2015). Why you are probably more creative (and less creative) than you think. In M. Karwowski, & J.C. Kaufman (Eds.), Creativity and the self. New York: Academic Press (in press). Baer, J., & Kaufman, J.C. (2008). Gender differences in creativity. Journal of Creative Behavior, 42, 75–105. Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman. Bandura, A. (2006). Guide for constructing self-efficacy scales. In F. Pajares, & T. Urdan (Eds.), Self-efficacy beliefs of adolescents (pp. 307–337). Greenwich, CT: Information Age. Batey, M., & Furnham, A. (2008). The relationship between measures of creativity and schizotypy. Personality and Individual Differences, 8, 816–821. Baudson, T.G., Fischbach, A., & Preckel, F. (2014). Teacher judgments as measures of children's cognitive ability: A multilevel analysis. Learning and Individual Differences. http://dx.doi.org/10.1016/j.lindif.2014.06.001. Begeny, J.C., Eckert, T.L., Montarello, S.A., & Storie, M.S. (2008). Teachers' perceptions of students' reading abilities: an examination of the relationship between teachers' judgments and students' performance across a continuum of rating methods. School Psychology Quarterly, 23, 43–55. Beghetto, R.A. (2006). Creative self-efficacy: correlates in middle and secondary students. Creativity Research Journal, 18, 447–457. Beghetto, R.A. (2014). Creative mortification: an initial exploration. Psychology of Aesthetics, Creativity, and the Arts, 8, 266–276. Beghetto, R.A., Kaufman, J.C., & Baxter, J. (2011). Answering the unexpected questions: student self-beliefs and teacher ratings of creativity in elementary math and science. Psychology of Aesthetics, Creativity, and the Arts, 5, 342–349. Bianco, M., Harris, B., Garrison-Wade, D., & Leech, N. (2011). Gifted girls: gender bias in gifted referrals. Roeper Review, 33, 170–181. Brandau, H., Daghofer, F., Hollerer, L., Kaschnitz, W., Kellner, K., Kirchmair, G., ... Schlagbauer, A. (2007). The relationship between creativity, teacher ratings on behavior, age, and gender in pupils from seven to ten years. Journal of Creative Behavior, 41, 91–113. Brown, T.A. (2015). Confirmatory factor analysis for applied research. New York: Guilford Press. Butz, A.R., & Usher, E.L. (2015). Salient sources of early adolescents' self- efficacy in two domains. Contemporary Educational Psychology. http://dx.doi.org/10.1016/j. cedpsych.2015.04.001. Carson, S.H., Peterson, J.B., & Higgins, D.M. (2005). Reliability, validity, and factor structure of the creative achievement questionnaire. Creativity Research Journal, 17, 37–50. Cervone, D., & Peake, P.K. (1986). Anchoring, efficacy, and action: the influence of judgmental heuristic on self-efficacy judgments and behavior. Journal of Personality and Social Psychology, 50, 492–501. Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155–159. Correll, S.J. (2001). Gender and the career choice process: the role of biased selfassessment. American Journal of Sociology, 101, 1691–1739. Darley, J.M., & Fazio, R.H. (1980). Expectancy confirmation processes arising in the social interaction sequence. American Psychologist, 35, 867–881. Dawson, V.L., D'Andrea, T., Affinito, R., & Westby, E.L. (1999). Predicting creative behavior: a reexamination of the divergence between traditional and teacher-defined concepts of creativity. Creativity Research Journal, 12, 57–66. Forgasz, H., & Leder, G. (2001). “A+ for girls, B for boys”: changing perspectives on gender equity and mathematics. In B. Arweh, H. Forgasz, & B. Nebres (Eds.), Sociocultural research on mathematics education: A international perspective (pp. 347–366). Mahwah, NJ: Lawrence Erlbaum. Freedman-Doan, C., Wigfield, A., Eccles, J., Blumenfeld, P.B., Arbreton, A., & Harold, R.D. (2000). What am I best at? Gender and grade differences in elementary school-age children's beliefs about their abilities at different activities. Applied Developmental Psychology, 21, 379–402. Furnham, A., Hosoe, T., & Tang, T. (2002). Male hubris and female humility? Acrosscultural study of ratings of self, parental and sibling multiple intelligence in America, Britain and Japan. Intelligence, 30, 101–115. Gabriel, S., & Gardner, W.L. (1999). Are there his” and “her” types of independence? The implications of gender differences in collective versus relational interdependence for affect, behavior, and cognition. Journal of Personality and Social Psychology, 77, 642–655.

7

Gentile, B., Grabe, S., Dolan-Pascoe, B., Twenge, J.M., Wells, B.E., & Maitino, A. (2009). Gender differences in domain-specific self-esteem: a meta-analysis. Review of General Psychology, 13, 34–45. Gilson, L.L., & Madjar, N. (2011). Radical and incremental creativity: antecedents and processes. Psychology of Aesthetics, Creativity, and the Arts, 5, 21–28. Glăveanu, V.P. (2010). Paradigms in the study of creativity: introducing the perspective of cultural psychology. New Ideas in Psychology, 28, 79–93. Glăveanu, V.P. (2014). The psychology of creativity: A critical reading. Creativity. Theories – Research – Applications, 1, 10–32. http://dx.doi.org/10.15290/ctra.2014.01.01.02. Glăveanu, V.P. (2015). The status of the social in creativity studies and the pitfalls of dichotomic thinking. Creativity. Theories – Research – Applications, 2, 102–119. http://dx.doi.org/10.1515/ctra-2015-0016. Gralewski, J., & Karwowski, M. (2013). Polite girls and creative boys? Students' gender moderates accuracy of teachers' ratings of creativity. Journal of Creative Behavior, 47, 290–304. Gralewski, J., & Karwowski, M. (2015). Are teachers' implicit theories of creativity related to recognition of their students' creativity? Journal of Creative Behavior (in press). Guilford, J.P. (1967). The nature of human intelligence. New York: McGraw-Hill. Harris, J.A. (2004). Measured intelligence, achievement, openness to experience, and creativity. Personality and Individual Differences, 36, 913–929. Helson, R. (1990). Creativity in women: inner and outer views over time. In M. Runco, & R. Albert (Eds.), Theories of creativity. Newbury Park: Sage. Hinnant, J.B., O'Brien, M., & Ghazarian, S.R. (2009). The longitudinal relations of teacher expectation to achievement in the early school years. Journal of Educational Psychology, 101, 662–670. Hoff, E.V., & Carlsson, I. (2011). Teachers are not always right: links between teacher ratings and students'' creativity scores, self-images, and self-ratings in school subjects. The Open Educational Journal, 4, 120–129. Hsu, L.A., Hou, S.T., & Fan, H.L. (2011). Creative self-efficacy and innovative behavior in a service setting: optimism as a moderator. Journal of Creative Behavior, 4, 258–272. Hughes, D.J., Furnham, A., & Batey, M. (2013). The structure of personality predictors of self-rated creativity. Thinking Skills and Creativity, 9, 76–84. Jacobs, J.E., Davis-Kean, P., Bleeker, M., Eccles, J.S., & Malanchuk, O. (2005). I can, but I don't want to: the impact of parents, interests, and activities on gender differences in math. In A. Gallagher, & J.C. Kaufman (Eds.), Gender differences in mathematics (pp. 246–263). Cambridge University Press. Jussim, L. (2015). Précis of social perception and social reality: why accuracy dominates bias and self-fulfilling prophecy. Behavioral and Brain Sciences. http://dx.doi.org/10. 1017/S0140525X1500062X (FirstView Article, in press). Jussim, L., & Harber, K.D. (2005). Teacher expectations and self-fulfilling prophecies: knowns and unknowns, resolved and unresolved controversies. Personality and Social Psychology Review, 9, 131–155. Karwowski, M. (2007). Teachers' nominations of students' creativity: should we believe them? Are the nominations valid? Social Sciences, 2, 264–269. Karwowski, M. (2009). I’m creative, but am I Creative? Similarities and differences between self-evaluated small and Big C creativity in Poland. International Journal of Creativity and Problem Solving, 19, 7–26. Karwowski, M. (2011). It doesn't hurt to ask… But sometimes it hurts to believe: Polish students' creative self-efficacy and its predictors. Psychology of Aesthetics, Creativity, and the Arts, 5, 154–164. Karwowski, M. (2012). Did curiosity kill the cat? Relationship between trait curiosity, creative self-efficacy and creative personal identity. Europe's Journal of Psychology, 8, 547–558. Karwowski, M. (2014). Creative mindset: measurement, correlates, consequences. Psychology of Aesthetics, Creativity, and the Arts, 8, 62–70. Karwowski, M. (2015a). Peers effect on students' creative self-concept. Journal of Creative Behavior, http://dx.doi.org/10.1002/jocb.102 (Article first published online: 7 SEP 2015). Karwowski, M. (2015b). Development of the creative self-concept. Creativity: TheoriesResearch-Applications, 2, 125–140. Karwowski, M. (2015c). Notes on creative potential and its measurement. Creativity. Theories – Research – Applications, 2, 4–16. Karwowski, M., & Barbot, B. (2015). Creative self-beliefs: Their nature, development and correlates. In J.C. Kaufman, & J. Baer (Eds.), The Cambridge Companion to Creativity and Reason in Cognitive Development (in press). Karwowski, M., & Lebuda, I. (2015a). The big five, the huge two and creative self-beliefs: A meta-analysis. Psychology of Aesthetics, Creativity, and the Arts (in press-a). Karwowski, M., & Lebuda, I. (2015b). creative self-concept: a surface characteristic of creative personality. In G. Feist, R. Reiter-Palmon, & J.C. Kaufman (Eds.), Handbook of creativity and personality. New York: Cambridge University Press (in press-b). Karwowski, M. & Szumski, G. (2015). Creative versus academic self-concept. Unpublished manuscript. Karwowski, M., Gralewski, J., Jankowska, D.M., Gajda, A., Wiśniewska, E., & Lebuda, I. (2015). Greater male variability in creativity: a longitudinal and cross-sectional latent variables approach. Journal of Creative Behavior (in press). Karwowski, M., Lebuda, I., Wisniewska, E., & Gralewski, J. (2013). Big five personality factors as the predictors of creative self-efficacy and creative personal identity: does gender matter? Journal of Creative Behavior, 47, 215–232. Kaufman, J.C. (2006). Self-reported differences in creativity by ethnicity and gender. Applied Cognitive Psychology, 20, 1065–1082. Kaufman, J.C. (2012). Self-estimates of general, crystallized, and fluid intelligences in an ethnically diverse population. Learning and Individual Differences, 22, 118–122. Kaufman, J.C. (2015). Creativity 101 (2nd edition ). New York: Springer (in press). Kaufman, J.C., & Beghetto, R.A. (2009). Beyond big and little: the four c model of creativity. Review of General Psychology, 13, 1–12.

8


Kirton, M.J. (1976). Adaptors and innovators. A description and measure. Journal of Applied Psychology, 61, 622–629. Kuklinski, M.R., & Weinstein, R.S. (2001). Classroom and developmental differences in a path model of teacher expectancy effects. Child Development, 72, 1554–1578. Lebuda, I. (2014). Kształtowanie poczucia sukcesu nauczycieli [the development of teachers' sense of success]. Kraków: Nomos. Li, Q. (1999). Teacher's beliefs and gender differences in mathematics: a review. Educational Research, 41, 63–76. Lim, H.S., & Choi, J.N. (2009). Testing an alternative relationship between individual and contextual predictors of creative performance. Social Behavior and Personality, 37, 117–136. Lindley, H.A., & Keithley, M.E. (1991). Gender expectations and student achievement. Roeper Review, 13, 213–215. Marsh, H.W., & Yeung, A.S. (1998). Longitudinal structural equation models of academic self-concept and achievement: gender differences in the development of math and English constructs. American Educational Research Journal, 35, 705–738. Meisels, S.J., Bickel, D.D., Nicholson, J., Xue, J., & Atkins-Burnett, S. (2001). Trusting teachers' judgments: a validity study of a curriculum-embedded performance assessment in kindergarten to grade 3. American Educational Research Journal, 38, 73–95. Piirto, J. (2004). Understanding creativity. Scottsdale, AZ: Great Potential Press. Plucker, J.A., Beghetto, R.A., & Dow, G.T. (2004). Why isn't creativity more important to educational psychologists? Potentials, pitfalls, and future directions in creativity research. Educational Psychologist, 39, 83–96. Pretz, J., & McCollum, V. (2014). Self-perceptions of creativity do not always reflect actual creative performance. Psychology of Aesthetics, Creativity, and the Arts, 8, 227–236. Raudenbush, S.W. (1984). Magnitude of teacher expectancy effects on pupils IQ as a function of the credibility of expectancy induction: a synthesis of findings from 18 experiments. Journal of Educational Psychology, 76, 85–97. Ready, D.D., & Wright, D.L. (2011). Accuracy and inaccuracy in teachers' perception of young children's cognitive abilities: the role of child background and classroom context. American Educational Research Journal, 48, 335–360. Renk, K., & Phares, V. (2004). Cross-informant ratings of social competence in children and adolescents. Clinical Psychology Review, 24, 239–254. Robinson, J.P., & Lubienski, S.T. (2011). The development of gender achievement gaps in mathematics and reading during elementary and middle school: examining direct cognitive assessments and teacher ratings. American Educational Research Journal, 48, 268–302. Rubie-Davies, C.M. (2006). Teacher expectations and student self-perceptions: exploring relationship. Psychology in the Schools, 43, 537–552.

Rudasil, K.M., Niehaus, K., Buhs, E., & White, J.M. (2013). Temperament in early childhood and peer interactions in third grade: the role of teacher-child relationships in early elementary grades. Journal of School Psychology, 51, 701–716. Rudowicz, E. (2004). Applicability of the ttest of creative thinking-–drawing production for assessing creative potential of Hong Kong adolescents. Gifted Child Quarterly, 48, 202–218. Silvia, P.J., Nusbaum, E.C., Berg, C., Martin, C., & O'Connor, A. (2009). Openness to experience, plasticity, and creativity: exploring lower-order, higher-order, and interactive effects. Journal of Research in Personality, 43, 1087–1090. Simpkins, S.D., Fredricks, J.A., & Eccles, J.S. (2012). Charting the eccles' expectancy-value model form mother's beliefs in childhood to youths' activities in adolescence. Developmental Psychology, 48, 1019–1032. Sommer, U., Fink, A., & Neubauer, A. (2008). Detection of high ability children by teachers and parents: psychometric quality of new rating checklists for the assessment of intellectual, creative and social ability. Psychology Science Quarterly, 50, 189–205. Sternberg, R., & Lubart, T. (1999). The concept of creativity: prospect and paradigm. In R. Sternberg (Ed.), Handbook of creativity (pp. 3–15). Cambridge: Cambridge University Press. Südkamp, A., Kaiser, J., & Möller, J. (2012). Accuracy of teachers' judgments of students' academic achievement: a meta-analysis. Journal of Educational Psychology, 104, 743–762. Tierney, P., & Farmer, S.M. (2002). Creative self-efficacy: its potential antecedents and relationship to creative performance. Academy of Management Journal, 45, 1137–1148. Tierney, P., & Farmer, S.M. (2004). The Pygmalion process and employee creativity. Journal of Management, 30, 413–432. Tierney, P., & Farmer, S.M. (2011). Creative self-efficacy development and creative performance over time. Journal of Applied Psychology, 96, 277–293. Urhahne, D. (2015). Teacher behavior as a mediator of the relationship between teacher judgment and students' motivation and emotion. Teaching and Teacher Education, 45, 73–82. Westby, E.L., & Dawson, V.L. (1995). Creativity: asset or burden in the classroom? Creativity Research Journal, 8, 1–10. von Wittich, D., & Antonakis, J. (2011). The KAI cognitive style inventory: was it personality all along? Personality and Individual Differences, 50, 1044–1049. Wolter, I., Gluer, M., & Hannover, B. (2014). Gender-typicality of activity offerings and child-teacher relationship closeness in German “kindergarten”. influences of the development of spelling competence as an indicator of early basic literacy in boys and girls. Learning and Individual Differences, 31, 59–65.