Attending to affect

Abstract

A bit more than 10 years after Alsop and Watts pointed out that "Despite the widespread belief that emotions are a central part of learning and teaching, contemporary work in science education exploring affect is scant" (2003, p. 1043), the level of attention given by science education researcher to affect has changed little. In the 11 years spanning 2001–2011, less than 10% of the articles published in the Journal of Research in Science Teaching (JRST), Science Education (SciEd), and the International Journal of Science Education (IJSE) have dealt with emotional perspectives on teaching and learning science, such as interest, motivation, attitudes, and self-efficacy, sometimes called affect (Alsop & Watts, 2003). While this 10% actually reflects a significant number of articles (138), when one considers the centrality of affect to teaching and learning and the broad range of topics that are related to affect, it is concerning that it has received relatively so little attention. With the hope of promoting awareness of the importance of this topic and past research on it, the rest of this article provides (A) my hypothesis why affect has been under-attended to by the science education research community and the ramifications of this under-attendance and (B) an overview of the research on affect in science education that has been published in JRST, SciEd, and IJSE between 2001 and 2011. I have made no attempt to synthesize or do a meta-analysis of this research; my purpose is to provide readers with a sense of some of the important work that has been done, to guide researchers and teachers to articles that may be relevant to their work, and to point out some weaknesses that should be avoided in the future. The overview ends by directing readers to a virtual issue of JRST on affect which presents some excellent examples of studies on affect that were published by JRST in the past decade. Why has relatively little research been published during the first decade of the 21st century on issues related to affect in science education? I surmise that this paucity of published research on affect in science education may be in part related to the international trend towards standardization of schooling and high-stakes testing. A search through the national education standards of two western countries (USA and Israel) indicated that not only are they overwhelmingly dominated by content- and practice-driven expectations, but affect is barely mentioned, and when it is mentioned, it is only dealt with superficially. For example, nowhere in the US national science education standards (National Research Council, 1996) are interest in science, motivation to engage with science in and out of school, positive attitudes toward and beliefs about science and the learning of science, or self-efficacy in science listed as one of the required outcomes of K-12 science education. Interest is mentioned in a few places as something that can support learning. The same is true for the new US Framework for K-12 Science Education (NRC, 2011) and Next Generation of Science Standards (Achieve, 2013). For schools to support students in learning the content and practice goals set by standards, the science education research community ought to increase attention to affect. Without motivation, interest, positive attitudes and self-efficacy, there can be only limited and curtailed engagement, and without engagement, learning is partial at best. "Because motivation leads to engagement, motivation is where teachers need to begin… without it, teachers have no point of entry" (Irvin, Meltzer, & Dukes, 2007, p. 32). Until we learn how to create schools, classes, and informal environments that continuously kindle and feed the desire to learn, we should not be surprised if we do not see much learning occurring. Until we, as a community of researchers and practitioners, identify positive affect as a crucial component of any educational system and value it enough so that it appears prominently in standards documents, we should not be surprised that many schools continue to be "boring places marked by drudgery and repetition where isolated students work in joyless and meaningless lessons painfully tied to their development level" (Kincheloe & Steinberg, 1999, p. 238). What is the point expecting that all students meet pre-specified learning goals by certain ages if at the same time we extinguish their curiosity, their desire to learn more, their belief in their ability use what they have learned, their trust that what they are learning is of importance? Positive affect should be as much a desired outcome of schooling as it is "a necessary condition for learning to occur" (Perrier & Nsengiyumva, 2003). Regardless of whether students will go on to work in a STEM-related profession or just live in a STEM-influenced world, we should strive for all to have positive attitudes to science and its role in society, motivation to understand the science of issues directly related to their lives and their general well-being, and a belief in their ability to make sense of issues. For example, when one is confronted simultaneously with magazine articles or TV programs that either support genetic engineering of food (Snell et al., 2012) or reject it (Dona & Arvanitoyannis, 2009), how is one to decide what to eat? If people are not to blindly follow the opinion they happened to read that day, they need to be able to make sense of the contrasting opinions and the evidence supporting them. This requires both a basic knowledge-base and the ability to learn new ideas, but more importantly, it requires a desire to understand the science concepts needed to make sense of the issue at hand and a belief in one's ability to learn these science concepts and apply them in reaching decisions, otherwise one will never go beyond accepting or ignoring the information provided by the TV program. Schools must support all students, regardless of where their main interests lie, in becoming life-long informed consumers and critics of STEM and its products. This can be achieved only by helping them construct not only a basic knowledge-base in science but also by igniting, maintaining, and reinforcing the desire to engage with science. "Education is not the filling of a pail but the lighting of a fire" (sometimes attributed to Yeats). Too often, however, research suggests that schools do just the opposite—they extinguish this fire (Osborne, Simon, & Collins, 2003; Yager & Penick, 1986). Typical 6-year old children are full of wonderment and curiosity about the world. By age 14, however, many youth are no longer inquisitive, and have lost their awe of nature, and their drive to understand and make sense of the world. The decline in students' attitudes toward, interest in, and motivation to learn science has been documented (e.g., Anderman & Young, 1994) and is apparent to many who teach science to early adolescents. Studies (Nolen, 2003; Vedder-Weiss & Fortus, 2011) have shown that this decline is not an inevitable consequence of adolescence, but that the school and class environment have a strong influence on how students position themselves in relation to science; schools and teachers can foster students' desire to engage with science or they can make it "something we learn because we have to, not because we want to" (my son). Understanding the reasons for this decline in affect toward science, how the various environmental factors such as parents, peers, teachers, schools, museums, internet, television, etc., interact with individuals to shape their affect in general and toward science in particular (NRC, 2009), should be one of the central goals of the science education community, and urgently so. Between 2001 and 2011, JRST published 51 articles on topics related to affect, SciEd published 27 articles, and IJSE 60. Given that each journal publishes a different number of issues and articles each year, the ratio between the number of articles on affect and the total number of articles per journal is very similar: a bit under 10%. I chose to focus on these three journals for three reasons: (A) they accept the broadest range of research specific to science education, (B) that are all ranked by ISI's social science index, and (C) when I asked my colleagues what they typically read to learn about the most up-to-date research in science education, these were the three journal that were consistently mentioned. An example of how studies on affect are under-represented in science education can be seen by following the process of submitting a manuscript to JRST, SciEd, and IJSE. Each of these journals requires the author(s) to select keywords that characterize the manuscript. In JRST and IJSE the authors choose from a pre-determined list of keywords while in SciEd there is no pre-specified list. In JRST the list contains 95 possible keywords of which only 1(!) is related to affect—"attitudes". In IJSE the list contains 142 possible keywords of which only 2(!) are related to affect—"affective domains" and "attitudes." There is no mention of motivation, interest, self-efficacy, self-concept… There was great variance between the types of articles published by each journal. I identified the methodological approach used in each study by reading through each manuscript's methods section and characterized it as being either qualitative, quantitative, mixed methods, or non-empirical (e.g., a review paper). In JRST, the ratio between articles that drew on a qualitative approach to those that drew on a quantitative approach to those that drew on mixed methods to those that were non-empirical was 38:7:3:0, respectively. The same ratio for SciEd was 16:7:2:1 and for IJSE 4:47:6:3. Thus, while IJSE provided a greater outlet for quantitative research on affect than the other journals, JRST was clearly the place for qualitative research on affect. Most of the published articles in these journals focused on attitudes (77), fewer on interest (39), fewer still on motivation (29), and even less on self-efficacy and self-concept (11). A number of articles were counted more than once since they dealt with multiples aspects of affect, such as the relation between self-efficacy and motivation. The research populations were also not evenly distributed: high-school (50) and middle-school students (33) were the main research populations, then elementary school students (20), university students that were not necessarily pre-service teachers (16), general adults (9), practicing teachers (8), pre-service teachers (7), and scientists (3). As Osborne et al. (2003) indicated in their review of research on attitudes in science education, there are primarily two kinds of attitudes of concern in science education—attitudes toward science, scientists, and science learning, and scientific attitudes which are more akin to habits of mind. In this overview I am concerned with only the first kind of attitude. A broad range of attitudes were studied: attitudes toward specific science courses (e.g., Cheung, 2009; Kitchen, Reeve, Bell, Sudweeks, & Bradshaw, 2007; Tien, Roth, & Kampmeier, 2002), research experiences (e.g., Seymour, Hunter, Laursen, & DeAntoni, 2004), science articles (Halkia & Mantzouridis, 2005; McClune & Jarman, 2010), learning science (e.g., Anders et al., 2003; Kind, Jones, & Barmby, 2007; Miller, Blessing, & Schwartz, 2006), teaching science (e.g., McGinnis et al., 2002; Pell & Jarvis, 2003), science content standards (Banilower, Heck, & Weiss, 2007; Donnelly & Boone, 2007), physics (e.g., Krogh & Thomsen, 2005; Lawrenz, Wood, Kirchhoff, Kim, & Eisenkraft, 2009; Zacharia, 2003), chemistry (e.g., Dalgety, Coll, & Jones, 2003; Salta & Tzougraki, 2004), the environment (e.g., Brossard, Lewenstein, & Bonney, 2005; Fernández-Manzanal, Rodríguez-Barreiro, & Carrasquer, 2007; Thompson & Mintzes, 2002), and science in general (e.g., Caleon & Subramaniam, 2005; Falk & Needham, 2011; Jarvis & Pell, 2005; Palmer, 2002; Rennie & Williams, 2002; Siegel & Ranney, 2003; Zacharia & Barton, 2004). There have also been articles that reviewed the literature on attitudes or specific instruments for measuring attitudes (e.g., Blalock et al., 2008; Lichtenstein et al., 2008; Osborne et al., 2003; Owen et al., 2008). The vast majority of all these studies were quantitative with only five being qualitative and another four drawing on both quantitative and qualitative methods. All the quantitative studies used surveys built around Likert-type items. Most of the studies developed their own instruments, sometimes adapting existing scales. A large fraction (25%) of all the articles on attitudes defined the development, critiquing, and reviewing of an instrument for measuring various aspects of attitudes as one of their primary goals. The development of so many different instruments for measuring various attitudinal aspects, often for very similar purposes, presents particular challenges. As a community we should strive to develop a number of standard instruments that can be adapted and translated according to researchers' specific needs. When each study uses a different instrument, it becomes very difficult to compare and synthesize findings. Developing high-quality instruments is a difficult and time-consuming task and it would be better for all if there was a compendium of accepted instruments to which one goes when wanting to measure an attitudinal aspect. Three of the surveys in these studies were developed using Rasch techniques (Donnelly & Boone, 2007; Scantlebury, Boone, Kahle, & Fraser, 2001; Siegel & Ranney, 2003). Since Likert-type items generate ordinal rather than interval data, any analysis of them using statistical procedures designed for interval data, such as t-tests, ANOVA and regressions, is suspect; Rasch analysis is required to transform ordinal data into interval data (Boone & Townsend, 2010). Thus, there is concern regarding the validity of many of the rest were not, raising concern about their validity. There was great variability in the design of the studies. Typical formats investigated changes in attitudes following an intervention (e.g., Gibson & Chase, 2002; Jarvis & Pell, 2002; Kanter & Konstantopoulos, 2010; Klop, Severiens, Knippels, van Mil, & Ten Dam, 2010; Luehmann & Markowitz, 2007; Scherz & Oren, 2006; Tomas, Ritchie, & Tones, 2011; Zacharia, 2003), changes in attitudes over time (e.g., Barmby, Kind, & Jones, 2008; Mattern & Schau, 2002; Siegel & Ranney, 2003), describing the attitudes of a sample and the relations of these attitudes to other constructs, such as religiosity, exploration processes, content knowledge, gender, teachers' decisions and practices, cultural border crossing, situational interest (e.g., Donnelly & Boone, 2007; Krogh & Thomsen, 2005; Lawrenz et al., 2009; Marshall & Young, 2006; Palmer, 2004; Thompson & Mintzes, 2002; Zint, 2002) or the factors that be influencing these attitudes (e.g., Raved & Assaraf, 2010; Reid & Skryabina, 2003; Scantlebury et al., 2001), and comparing attitudes of differing populations, sometimes due to differing treatments (e.g., Bybee & McCrae, 2011; Cavallo & Laubach, 2001; Gwimbi & Monk, 2003; Jones et al., 2004; Machina & Gokhale, 2010; Rennie & Williams, 2002; Tien et al., 2002). Once they have formed, attitudes tend to be inert and slow to change (Ajzen & Fishbein, 1980). As mentioned above, several of these studies investigated how attitudes change following an intervention. Given that many of these interventions were relatively short, often lasting only a few hours, it is surprising that several of these studies identified significant changes to their participants' attitudes. Is it possible that issues with the validity of the instruments used (see above) led to type II errors? Or perhaps these attitude changes were temporary and reversed once the intervention faded in the past? There are several mainstream motivational theories that are in use today (Schunk, Pintrich, & Meece, 2007): achievement goal theory (Ames, 1992; Ames & Archer, 1988), expectancy-value theory (Eccles & Wigfield, 2002; Eccles et al., 1983; Wigfield, Eccles, & Rodriguez, 1998), attribution theory (Weiner, 2005), self-determination theory (Deci, Vallerand, Pelletier, & Ryan, 1991), and social cognitive theory (Bandura, 2001). Most of the motivation studies published by JRST, SciEd, and IJSE between 2001 and 2011 drew on social cognitive theory (e.g., Bryan, Glynn, & Kittleson, 2011; Glynn, Taasoobshirazi, & Brickman, 2009; Palmer, 2005); others used an achievement goal perspective (e.g., Nolen, 2003; Vedder-Weiss & Fortus, 2011; Velayutham, Aldridge, & Fraser, 2011), and one used the lens of expectancy value theory (Patrick, Mantzicopoulos, & Samarapungavan, 2009). Five studies used none of the conventional motivation theories to frame themselves; they used the term motivation without providing a theoretical framing for the term. A few used a unique theoretical framework for motivation in museum-based learning (e.g., Falk & Storksdieck, 2005; Kisiel, 2005). Twenty-three of the 29 articles on motivation used quantitative methods. Most used surveys based on Likert-type items (e.g., Glynn, Brickman, Armstrong, & Taasoobshirazi, 2011; Juriševič, Glažar, Pučko, & Devetak, 2008; Nolen, 2003; Vedder-Weiss & Fortus, 2011), though there were also studies that quantitatively analyzed interviews (Patrick et al., 2009; Venturini, 2007). Four articles were based on qualitative methods and used primarily interviews (Falk & Storksdieck, 2010; Russell & Atwater, 2005), interviews and field-notes (Abrahams, 2009), and questionnaires with open-ended questions (Kisiel, 2005). There was 1 mixed-methods study (Stake & Mares, 2005) that used interviews that were analyzed both quantitatively and qualitatively. Finally, one article was theoretical and reported no empirical data (Palmer, 2005). Studies dealt with learning in informal settings (Falk & Storksdieck, 2005, 2010; Kisiel, 2005), the future intents of the participants (Russell & Atwater, 2005), identifying time, topic, school, instructional design, brain-type and gender-based trends (Abrahams, 2009; Bryan et al., 2011; Juriševič et al., 2008; Vedder-Weiss & Fortus, 2011; Zeyer & Wolf, 2010), outcomes of interventions (Berger & Hänze, 2009; Patrick et al., 2009; Stake & Mares, 2005; Sturm & Bogner, 2008), the relations between motivation and achievement (Devetak & Glažar, 2010; Glynn, Taasoobshirazi, & Brickman, 2007; McLellan, 2006; Nolen, 2003; Saçkes, Trundle, Bell, & O'Connell, 2011; Tseng, Tuan, & Chin, 2010; Wang, Wu, & Huang, 2007; Zusho, Pintrich, & Coppola, 2003), the development of instruments (Glynn et al., 2009, 2011; Tuan, Chin, & Shieh, 2005; Velayutham et al., 2011), the development of a motivational model of teaching (Palmer, 2005), and an alternative to traditional motivation theories (Venturini, 2007). As mentioned earlier, academic work on interest has focused primarily on two directions: (A) the characteristics of an object, idea or event that make it more or less interesting to many people (Todt, Drewes, & Heils, 1994), and (B) the characteristics of the psychological of being in a class of or ideas, how interest may and how interest can influence motivation, engagement, and learning & 2002; & & 2006; & of the articles on interest that were published by JRST, SciEd, and IJSE between 2001 and 2011 not on any of these theoretical not what they by interest and used the term in an even though are of interest that from both the use of interest and a of differing research et al., p. A few of them mentioned such as interest or interest, but not any theoretical support for these The instruments these studies used built on an rather than of interest, so it is difficult to that they were measuring The articles that accepted theoretical on interest research on a range of the topics in which students are as by for and and by the questions they of (e.g., & 2009; & 2011; & 2011; & 2011), interest in physics and related to physics (e.g., & 2004; & 2010), situational interest in lessons and during a (e.g., 2011; Palmer, 2009), the relations between interest, learning and science (e.g., et al., 2003; & Huang, 2011), and an overview of research on interest in science Most of the studies were quantitative with 11 being qualitative and another using mixed methods. The quantitative data from three primary a of questions students have asked of scientists, and The of the quantitative analysis procedures was Most of the studies developed their own instruments, adapting existing drawing from the literature and the In a few there was no of the items or how they were so the of these studies are others provided no information on their not a analysis even though they many but to each many an of the process of the that their were and (e.g., & 2009; Falk & 2003; Palmer, a few also used Rasch as required on Likert-type data (e.g., et al., 2011; & Subramaniam, data was typically from interviews (e.g., & Barton, 2007; & Collins, 2001; & and (e.g., 2011; 2005), but also from et al., 2003). Studies positive between interest in science (A) a range of different achievement (e.g., & 2008; Falk & 2003; & 2009; et al., 2011; 2007), (B) 2009), (C) self-concept (e.g., 2007), and and future in science (e.g., & 2011; et al., 2010; & identified that students' situational social and 2011; Palmer, 2009), issues related to et al., 2009; & 2002; & 2010; et al., 2008), between countries & 2011), and the influence of schools and self-concept guide motivation in several influence the to engage in some and also affect how much people will on an or an how they will when and how they will be in of (Bandura, has consistently been shown to be one of the of motivation and (Schunk, articles on self-efficacy and self-concept were published by JRST, SciEd, and IJSE between 2001 and All were based on quantitative methods, using questionnaires and data, for two & 2007; et al., that used both questionnaires and interviews and as their data All the questionnaires were based on Likert-type items that were and adapted from existing for the study by and that used a and the study by which used both items and items from an existing studies that self-efficacy and self-concept are with a range of in chemistry 2007), ability to make between and science experiences 2010), in science 2008; Bryan et al., 2011; & 2002) and on & 2009). In while self-efficacy was with 2009), no significant relation was between the of science teachers in a school and students' science self-concept et al., 2007), and self-concept was to a more important role during the of an intervention than during the practice et al., 2003). While a of research on affect has been published in the past decade and many important and this important topic under-attended to, when its on all aspects of science education. Understanding is but so is wanting to go and of one will The of any significant to affect in the Next Generation Science Standards and the very of published research in science education with affect that for many of affect and is as a central issue in science education. In an attempt to this of the of JRST to a virtual issue some of the excellent studies that dealt with affect in science education and were published by JRST between 2001 and 2011. articles were and the articles were to be published in a virtual issue of JRST on The 51 articles were identified by the of all the articles published by JRST in this for the following motivation, affect, interest, self-efficacy, and 51 articles were to a of which was of some of the authors of the 51 identified Each read five articles and them on three their interest to science teachers, and whether they would them to their to Coll, Glynn, Lawrenz, Nolen, Palmer, Saçkes, and Zacharia for excellent and many into the a range of factors (Nolen, 2003), youth & Barton, 2007), beliefs & an (Stake & Mares, 2005), 2004), an intervention with (Patrick et al., 2009), STEM & Roth, 2010), and school & Fortus, I hope that this virtual issue will be they science teachers or of the importance of affect in their work, of some of the excellent work that has been published by the science education community on this topic in the past and provide them with on to this important topic in the future.