Teacher training in computational thinking (CT) is becoming more and more important, as many countries are introducing CT at all K-12 school levels. Introductory programming courses are known to be di cult, and some studies suggest they foster an entity theory of intelligence ( xed mindset), reinforcing the idea that only some people have socalled \geek gene". This is particularly dangerous if thought by future primary school teachers. We analyzed the e ects of an introductory course about computational thinking and creative computing with Scratch, and observed a statistically signi cant increase of pre-service teachers' growth mindset while observing a statistically signi cant decrease in their computer anxiety. The structure of the course is detailed, with particular emphasis on some characteristics that may have determined growth mindset increase. Limitations of this exploratory study are discussed, and future work is depicted.
1 In the last decade, computational thinking (CT) has been recognized as a fundamental skill for everyone, not just computer scientists [Win06]. Many countries in the world are making e orts to include it in the school curriculum, at all K-12 levels [GP13].
In Italy, the recent school system reform explicitly states that it is mandatory to develop student's digital skills, with particular care to the development of computational thinking.
These pushes to teach CT (that very often is realized by teaching programming - or \coding") give rise to the necessity of an urgent plan for teacher's training, both for pre-service and in-service ones, and especially for primary school teachers. In Italy, in facts, primary teachers are not trained to teach CS fundamentals (and only since 2002 they need a Primary education degree to teach). Moreover, Italian primary teachers are mostly female, and therefore possibly subject to stereotypes about women and CS.
To non-computer scientists, learning to program may appear as a too challenging goal, achievable only from those having the so-called \geek gene" [AL13, PBCE16]. Moreover, stereotypes lead some people to identify computer scientists with singularly focused, asocial, competitive, male gures [LAY16].
Students and teachers have di erent personal ideas (\implicit theories") about their intellectual abilities. Some believe that their intelligence is a xed trait (like eye color or height when adult), and they cannot do much to change it: they have an entity theory of intelligence, otherwise stated a xed mindset. Some others believe instead that intelligence can be developed with study and e ort (like muscles can be trained): they have an incremental theory of intelligence, also called a growth mindset. Mindset theory is a fundamental result of Carol Dweck's research [Dwe00]. In many studies, she showed that student's mindset could predict their achievement, in particular in Math and Science, and their ability to cope with challenges [BTD07, Dwe08]. Moreover, female students with a growth mindset showed less susceptibility to the harmful e ects of stereotypes about women and math [GRD12]. In [MT08] it is argued that growth mindset can be particularly important in CS education. Teachers' growth mindset is strongly necessary (although not su cient) to foster a growth mindset in their students: teachers must create a growth mindset environment where growth messages are sent, but also teach students new strategies to cope with failures and to master the material [Dwe17].
Since 2014, at the University of Bologna, a laboratory course on \computational thinking and creative computing with Scratch" has been taught to preservice primary teachers. During the rst two years of teaching, instructors collected oral reports from students. At the beginning of the course, many of them reported anxiety and low self-e cacy about learning to program. Some of them described themselves as \not a computer science/technology person". On the contrary, at the end of the course, instructors received very good feedback. Some students spontaneously thanked them because they did not feel any more like they were not \computer science people" and felt empowered to be creative with technology and to teach it to their future pupils.
Based on these anecdotal pieces of evidence, we decided to explore changes in mindset and computer anxiety before and after the fall term course of the academic year 2016-2017. The test on mindset was replicated in the spring term course. 2 2.1
rst used in 1980 by Seymour Papert in Mindstorms [Pap80] and then brought to the attention of our community by Jeannette Wing [Win06] in 2006. In this decade, a body of literature has been produced to search for a better de nition of this concept, and to provide tools and frameworks to introduce and assess CT in K-12 education [GP13]. While there is no agreement between authors, a lot of proposed de nitions recognize CT is not only about technical methods and practices, but also about mental processes and transversal skills like creativity, collaboration, tolerance for ambiguity, resilience, and more [CLN17]. 2.2
Scratch is a visual programming environment to create interactive media-rich projects, like video games, interactive stories, interactive art, and so on [MRR+10]. Scratch was developed at MIT Media Lab by the Lifelong Kindergarten group. It is built on constructionist ideas of Papert's LOGO [Pap80]. Some key features of Scratch, relevant for this work, are: liveness - the program is constantly running, and its behavior changes immediately when parts of the code are added, edited or removed; tinkerability - the program lets you experiment with blocks, in a very bottom-up, trial and error approach; no error messages - like when you play with LEGO R bricks, either blocks snaps together, or they don't; moreover, if your program is not correct, it runs anyway, so you don't feel too frustrated, and then you can try to gure out why it doesn't behave as expected.
Scratch's main goal is to teach digital uency as a means of self-expression rather than as a tool for future careers [RM+09]: often students can use technology as passive users, but few of them have the opportunity to be active creators through technology.
Scratch belongs to MIT's vision about creative learning, and was speci cally designed to help young people grow up as creative thinkers (through the \creative learning spiral" - an iterative approach to creativity - and through four main ingredients: Project, Peers, Passion, Play [Res14]).
These ideas are implemented in the Harvard's \Creative Computing Online Workshop"2 and \Scratch curriculum guide" [BBC14] and in the \Learning Creative Learning" course at MIT3. Materials from these initiatives represent the primary sources for the course presented in this paper. 2.3
Dweck's studies on growth mindset are based on three decades of research [Dwe00]. Students with growth mindset show learning-oriented goals (not afraid to ask questions and make mistakes, in order to learn) and mastery-oriented responses (greater e ort and new strategies) to challenges and setbacks, while students with xed mindset show performance goals (\appear intelligent", so avoiding di cult tasks) and helpless response to challenges (e.g. giving up or blaming the teacher for their failure). As said in Section 1, growth mindset is positively correlated with grades and achievements and can be useful to reduce gender disparities in STEM.
Growth mindset can be positively conveyed by some interventions [Dwe08]: explicitly teaching students about mindsets, brain plasticity and the idea that intelligence can be trained with e ort; portraying challenges, e ort and mistakes as highly valued; praising process and e ort, and give constructive feedback rather than praising the person or being judgmental.
Speci c suggestions to stimulate a growth mindset in Math includes also [Boa13]: giving rich open tasks, 2https://creative-computing.appspot.com/preview 3http://learn.media.mit.edu/lcl/ oriented to learning, requiring e ort and reasoning; teaching for patterns and connections; teaching creative and visual Mathematics.
Teachers' conceptions are crucial: in a study described in [Dwe08], adults were asked to behave as teachers - after a xed or a growth mindset about Math had been taught to them. The \growth" group was more supportive with students, giving encouragement and suggesting positive strategies to deal with problems; by contrast, the \ xed" group subjects gave students simple comfort and xed messages (e.g., \Not everyone is a math person! ") and helped boys signi cantly more than they did with girls. 2.4
Computer anxiety can be de ned as \a fear of computers when using one, or fearing the possibility of using a computer " [SON05], and di ers from negative attitudes toward computers. In facts, it involves a more a ective response: \resistance to and avoidance of computer technology are a function of fear and apprehension, intimidation, hostility, and worries that one will be embarrassed, look stupid, or even damage the equipment " [HGK87].
Computer anxiety has been correlated with math anxiety and gender [HGK87, Mau94]. Females were found to have higher computer anxiety. By contrast, previous exposure to computers is correlated with the low level of anxiety. As [Mau94] suggests, females have less exposure to computers than males due to stereotypes, so previous exposure should be taken into account.
A study, referenced by Dweck herself, correlates computer anxiety and self-theories. It showed that computer anxiety decreased in participants of a basic computer training course who were taught incremental conceptions of ability, while did not change in participants to whom xed entity conceptions of ability were taught [Mar94].
To assess computer anxiety, the \Computer Anxiety Rating Scale (CARS)" was developed and validated in [HGK87]. 3
As opposed to other scienti c disciplines, only a few studies have been conducted on the relationship between an introductory computer science/programming course and a growth mindset. In a survey administered to CS faculty members of a U.S. institution [Lew07], more than three-quarters of them disagreed on the fact that \Nearly everyone is capable of succeeding in the computer science curriculum if they work at it". Carol Dweck herself describes computer science as a discipline that requires a growth mindset [CCD+10].
Like math, computer science can induce a xed mindset, as some authors [MT08, CCD+10] suggest. By contrast, we think some intrinsic characteristics of CS/CT (at least if taught as a creative subject - e.g., open/real/authentic projects, iterative approach, debug, trial and error, collaboration rather than competition) can foster a growth mindset. In other elds, for example Engineering, similar results were found: during the rst year of University, students tend to move towards a xed mindset. However, introducing openended engineering design projects into the curriculum may tend to lessen or eliminate the shift toward xed mindset [RF14].
Only a few studies have been conducted to assess or alter the student's mindset before and after a programming course.
Simon et al. [SHM+08] tried a small intervention in CS1 classes to change the mindset of students from CS Majors and Minors, but they obtained mixed results.
Cutts et al. [CCD+10] performed three structured interventions into an introductory programming course, gaining signi cant improvement in growth mindset level of students and also a positive correlation in their test scores.
On the contrary, Flanigan et al. [FPSS15] analyzed (without intervention) changes in CS1 (CSmajor, other STEM-Majors, but also Arts and Business Majors) students across the semester, nding a signi cant increase in a xed mindset and a signi cant decrease in a growth mindset.
All the cited experiments were conducted among CS1 students. Authors of the present paper did not nd any study investigating correlations between growth mindset and CT courses. 4
We decided to observe growth mindset and computer anxiety changes between the beginning and the end of the laboratory course.
We decided to not teach explicitly about growth mindset or brain growth, and the instructor (a computer scientist with a background in education) paid particular attention in avoiding explicit mentioning of Dweck's research and ideas in lessons/suggested reading material to avoid in uencing the subjects and to test if CT and creative learning could in uence growth mindset.
No active intervention was made to in uence computer anxiety. Currently, to become a pre-school and/or a primary school teacher in Italy, you have to get a 5-year (Single cycle/Combined Bachelor and Master) Degree in
dents also get a \Pre-school and Primary school teaching license" that allows them to teach in Italian public schools. For historical and sociological reasons, in Italy primary teachers are mainly female and this is re ected in the fact that Primary teacher education students are almost all female (for instance, 91% in a.y. 2016/17 in our University).
At the University of Bologna, primary teacher education students take a mandatory \General Education and Educational Technologies" exam during the rst year, and follow a practical 24 hours (3 credits) \Educational Technology Laboratory" during the fourth year. For that course, they can choose between several topics, from the use of interactive white-board, to stop-motion storytelling techniques and many others, all with the aim to learn how to use technology as a tool for better teaching. To allow students to be supported by instructors and to work with technologies actively, each thematic-laboratory has a maximum of 32 students. In the context of this \multi-track course," since the academic year 2014-15, students can choose the \Laboratory of creative computing and computational thinking", to learn the basics of computational thinking and creative computing with Scratch. To give the opportunity to take the course to more students (of the same academic year), the laboratory is replicated (same instructor, schedule, location, contents) in the fall term and in the spring term. 4.2
The course was designed as an introduction to creative computing and computational thinking with Scratch. It was made up of 6 lessons, 4-hours each. The course plan is now described. Many activities are taken from MIT/Harvard materials (see sec. 2.2).
Brief introduction to creative computing and computational thinking; experiments with Google Presentations: the teacher creates a shared presentation with writing rights, then she asks students to add a new page and to write something about themselves, putting on a photo, and so on. This activity is initially messy, but soon students learn in a very bottom-up fashion to use the tool and to avoid modifying peers content; free exploration of Scratch; mini challenge to make something happen with it (\Scratch surprise" [BBC14]); guided tutorial to create a simple video game that contains a lot of computational concepts4.
Ten blocks challenge5; free artistic project with just the simple hint to use the \pen" and \looks" categories blocks; debug exercises: students had to choose some debug exercises from [BBC14], remix them, nd the bug and comment out how they found it and what they did to correct it.
Witness from an invited primary school teacher using unplugged activities and Scratch in her teaching; examples of Scratch projects to be used with pupils6; \about me" [BBC14] free project: create a project to introduce yourself and the things you do and love.
Exploration of Code.org and comparison with Scratch: try some activities, nd out pros and cons of the di erent platforms and approaches; advanced features (cloning and webcam): try to reproduce a \snowing-like" behavior with cloning and catching the clones with the hand through webcam-sensing7.
Scratch and the physical world (Makey Makey8 and Lego WeDo9) demos; time to work in small groups on the nal project. 4A simpli ed version of Carmelo Presicce's \Under the sea": https://scratch.mit.edu/projects/14759947/
5https://creative-computing.appspot.com/unit?unit=4& lesson=13 6e.g. from https://scratch.mit.edu/studios/1918506/ 7e.g. https://scratch.mit.edu/projects/129283065/ 8e.g. the classical \human chain" and \whack a mole": https://scratch.mit.edu/projects/43681296/
9e.g. simple sensor/motor use with Scratch described in: https://www.youtube.com/watch?v=qBhIcb-Ipmw
Public presentation of nal projects: design of an activity with Scratch for Primary School in the light of creative learning's 4 Ps. Students were advised to be particularly careful to not create a game or a project on which their pupils would have been passive consumers, but rather design a creative activity where their pupils should be free to create di erent projects but in the context of some Primary school teaching objectives of one or more subjects.
Each student had her own PC, but they were allowed to work in pairs/small groups, to get up and move around the laboratory and to communicate with each other. The teacher was always available to o er help.
The laboratory was mainly hands-on: students were assigned projects with a broad theme (e.g., \a project about you") and given time to freely create with Scratch, experimenting, getting help online or asking the instructor. Sometimes, more structured exercises were given, but they were chosen not to be mechanical or repetitive. Instead, they were explicitly posed as challenges to have fun with, while learning. No theoretical lectures about programming or Scratch were given. However, some tips and quick demos, always after they worked a while on projects or problems, were given.
Homeworks consisted in realizing other projects at home and writing a page in a shared online notebook (Google Presentation), re ecting on di culties, achievements, and learning process. The instructor gave feedback as comments, and students were invited to comment at least two of their mate's pages each week.
An online virtual class was set up with a Google+ community, where students could ask for help (to mates and to the teacher) and discuss. The instructor posted interesting videos/articles and stimulated comments.
The exam was pass/fail, with no grades. At the beginning of the course, it was clearly stated that students would have been evaluated for their e ort (measured with the presence in class, participation, shared projects) rather than on the quality of their works. The nal presentation of a group project was also mandatory to pass the exam. Students were in particularly encouraged to share their projects even if they were buggy or incomplete, and ask for help.
All students passed the exam in both terms. Projects and journals were not graded but were checked by the instructor, and written feedback was given.
No explicit reference to growth mindset and brain growth theories were made. However, other growth mindset strategies (see 2.3) were put in action: in particular it was carefully paid attention to give growth mindset feedback (both oral and written in the comments to projects or posts) and it was praised process and outcome (\You worked a lot to create that!", \Very good project!") rather than the person (\Bravo!", \You are very good at it!").
The instructor established a good class climate, where errors were not stigmatized but seen as a powerful tool for learning, helping students re ect on them with guided questions rather than simply \tell the solution". This was eased by the tool: Scratch helps you not be frustrated by errors and instead motivates you to gure out how to x your bugs.
Scratch tinkerability helped also to encourage students to not give up, moving forward by trial and error and feeling empowered by their learning and successes. 4.3
An identical survey was administered at the beginning (pre-survey) and at the end of the course (postsurvey). It was dived into two sections (Growth Mindset and Anxiety) in the fall term course, while had only the Growth Mindset section in spring term's one.
The rst section was intended to assess students' mindset through the Implicit Theories of Intelligence Scale (see Appendix A.1 for the full scale). It included eight Likert-type items, described in [Dwe00, p. 285]. Students were asked to rate from 1 (\completely disagree") to 6 (\completely agree") eight statements about intelligence (in Italian in the survey), four reecting an incremental theory, like \No matter who you are, you can signi cantly change your intelligence level " and four re ecting an entity theory, like \You have a certain amount of intelligence, and you can't really do much to change it ". During the analysis, the latter were reverse scored, so that high points were associated with a growth mindset, while low points with a xed mindset.
The second section was intended to assess student's anxiety through the Computer Anxiety Rating Scale (see Appendix A.2 for the full scale). It included nineteen Likert-type items, described in [HGK87] (we used an Italian translation of the slight variation [SON05] of the original statements). Students were asked to rate from 1 (\completely disagree") to 5 (\completely agree") proposed statements (in Italian in the survey), half of them re ecting a high anxiety (like \I have avoided computers because they are unfamiliar and somewhat intimidating to me" or \I do not think I would be able to learn a computer programming language") while half of them re ecting a low level of anxiety (like \Learning to operate computers is like learning any new skill, the more you practice, the better you become"). During the analysis, the latter were reverse scored, so that high points were associated with high anxiety, while low points with low anxiety.
The pre-survey was administered right at the beginning of the fall term course, when students knew only the title and a very brief description of the course from the website. The post-survey was administered at the end of the course. Five weeks passed between the two administrations.
The questionnaires were anonymous, but answers of the same subject to pre and post-questionnaire were linked with a randomly generated code unknown to the researcher. The questionnaires were administered with an online form (Google Form).
This process was repeated for the spring term course, but with mindset questions only.
A total of twenty-three students (N = 23), all females, aged from 21 to 29 (M = 23, SD = 1:95, M ODE = 22) completed both the pre-survey and the post-survey of the fall term course.
Moreover, a total of twenty students (N = 20), all females, aged from 22 to 28 (M = 23:05, SD = 1:82, M ODE = 22) completed both the pre-survey and the post-survey of the spring term course. 4.4
The data were analyzed with the R programming language and RStudio environment. 4.4.1
For each subject, the initial and nal growth mindset level was calculated. Growth mindset level is a value from 1 ( xed mindset) to 6 (growth mindset), calculated as the mean of the eight answers (with entity items reverse scored, as stated) of each subject.
A paired-samples t-test was conducted to compare growth mindset at the beginning and at the end of the fall term course. There was a statistically signi cant di erence in the mindset scores between the pre-test (M = 4:62, SD = 0:78, = 0:86) and the post-test (M = 4:90, SD = 0:76, = 0:86): t(22) = 2:35, p = 0:028 (< 0:05). In particular, growth mindset has increased from the beginning to the end of the course. (see Fig. 1, where mean GM of all subjects is represented as a black diamond).
Moreover, a paired-samples t-test was conducted to compare growth mindset at the beginning and at the end of the spring term course. Again, we found a statistically signi cant increase in the mindset scores between the pre-test (M = 4:08, SD = 0:80, = 0:76) and the post-test (M = 4:44, SD = 0:83, = 0:90): t(19) = 2:50, p = 0:022 (< 0:05) (see Fig. 1). For each subject, initial and nal computer anxiety level was calculated. Anxiety level is a value from 1 (low anxiety) to 5 (high anxiety), calculated as the mean of the nineteen answers (with some items reverse scored, as stated) of each subject.
A paired-samples t-test was conducted to compare computer anxiety at the beginning and at the end of the course. There was a statistically signi cant di erence in the anxiety levels at the beginning (M = 2:04, SD = 0:58, = 0:90) and at the end (M = 1:85, SD = 0:60, = 0:92): t(22) = 2:98, p = 0:007 (< 0:01). In particular, computer anxiety has decreased from the beginning to the end of the course (see Fig. 2). All answers showed an high internal consistency (see Cronbach's alphas in data analysis).
To be valid for a paired t-test, distribution of the di erences between the two related values of each subject should be approximately normally distributed. Di erences from both growth mindset and anxiety scores passed the Shapiro-Wilk normality test, recommended for small samples.
Finally, both measures are resistant to testretest [HGK87, DCH95]. 4.6
Since it is a pre-experimental design, this study is aficted by some limitations. In particular: there is no control group, and so we don't know if the course is the only or the main cause of the di erence between results, or if external factors may have intervened; as a positive observation, anyway, both fall and spring groups registered an increase in mindset; the sample was relatively small and not randomized: it was made up of students that decided to attend the class; changes may be in uenced by the experience of taking the test itself; regression towards the mean may have in uenced the results.
Despite the limitations of the study, we found a statistically signi cant increase in participants' growth mindset (result replicated in a following identical course) and a statistically signi cant decrease in participants' computer anxiety.
Initial growth mindset was already high. This is hardly surprising because it is crucial for teachers not to have xed views about intelligence and almost certainly this has been taught them in the previous years of their degree. However, this clearly contrasts with oral reports about their \CS con dence" collected at the beginning of the course. We suspect they may have a high growth mindset in general, but hold xed ideas about CS in particular: it is known one can have di erent mindsets in di erent areas [Dwe17].
More surprising is their medium/low initial anxiety. This may be due to their young age: they grew up in a world where technologies are everywhere, so they rapidly get used to them. It may be the case to construct an updated anxiety scale, which considers this di usion and tests anxiety about more profound skills related to computer science rather than computers themselves or use speci c CT/CS anxiety scales. It would also be interesting to measure CS self-e cacy.
Most interventions taught explicitly about brain growth to in uence self-theories about intelligence. Even though we recognize this is crucial, we aim to foster a growth mindset mainly with teaching innovations and fundamental aspects of computer science. \Explicitly teaching" interventions can be further positive boosters for a growth mindset.
The aim of our exploratory study was mainly to test whether our insights about CT, creative computing and growth mindset were correct. Now we have to: design a proper experiment to con rm these preliminary data; investigate more deeply what factors of CS/ CT/ creative computing are the most signi cant to foster growth mindset: we believe that teacher's feedback, iterative approach, open projects and challenging exercises were crucial; de ne and investigate speci c \computer science mindset", rather than general ideas about intelligence; evaluate the relationship between (CS) growth mindset and actual learning of CT concepts.
Many thanks to all the students of the courses; to Simone Martini, for supporting my research and for his useful suggestions; to Alessandro, for the help with statistics; to Carmelo and all CoderDojo friends, for making me discover Scratch and Creative learning.
A. Ahadi and R. Lister. Geek genes, prior knowledge, stumbling points and learning edge momentum: Parts of the one elephant? In Proc. of ICER 2013, pages 123{ 128, New York, NY, USA, 2013. ACM.
K. Brennan, C. Balch, and M. Chung.
Scratch curriculum guide, 2014. http://scratched.gse.harvard.edu/guide/.
J. Boaler. Ability and mathematics: the mindset revolution that is reshaping education. In Forum, volume 55, pages 143{ 152. Symposium Journals, 2013.
L. S. Blackwell, K. H. Trzesniewski, and C. S. Dweck. Implicit theories of intelligence predict achievement across an adolescent transition: A longitudinal study and an intervention. Child Development, 78(1):246{263, Jan 2007. [AL13] [BBC14] [Boa13] [BTD07] [CLN17]
I. Corradini, M. Lodi, and E. Nardelli.
Conceptions and misconceptions about computational thinking among italian primary school teachers. In Proc. of ICER 2017, pages 136{144, New York, NY, USA, 2017. ACM.
C. S. Dweck, C. Chiu, and Y. Hong. Implicit theories and their role in judgments and reactions: A word from two perspectives. Psychological Inquiry, 6(4):267{285, 1995.
Psychology Press, 2000.
C. S. Dweck. Mindsets and math/science achievement. The Opportunity Equation, 2008.
Robinson, 2017.
A. E. Flanigan, M. S. Peteranetz, D. F.
Shell, and L. Soh. Exploring changes in computer science students' implicit theories of intelligence across the semester. In Proc. of ICER 2015, pages 161{168, New York, NY, USA, 2015. ACM.
S. Grover and R. Pea. Computational Thinking in K-12: A Review of the State of the Field. Educational Researcher, 42(1):38{43, Jan 2013.
C. Good, A. Rattan, and C. S. Dweck.
Why do women opt out? Sense of belonging and women's representation in mathematics. Journal of Personality and Social Psychology, 102(4):700{717, 2012.
R. K. Heinssen, C. R. Glass, and L. A.
Knight. Assessing computer anxiety: Development and validation of the computer anxiety rating scale. Computers in Human Behavior, 3(1):49{59, Jan 1987.
C. M. Lewis, R. E. Anderson, and K. Yasuhara. "I Don'T Code All Day": Fitting in Computer Science When the Stereotypes Don'T Fit. In Proc. of ICER 2016, [Lew07] [Lod17] [Mar94] [Mau94] pages 23{32, New York, NY, USA, 2016.
ACM.
C. Lewis. Attitudes and beliefs about computer science among students and faculty. ACM SIGCSE Bulletin, 39(2):37, Jun 2007.
M. Lodi. Growth mindset in computational thinking teaching and teacher training. In Proc. of ICER 2017, pages 281{282, New York, NY, USA, 2017. ACM.
J. J. Martocchio. E ects of conceptions of ability on anxiety, self-e cacy, and learning in training. Journal of Applied Psychology, 79(6):819{825, 1994.
M. M. Maurer. Computer anxiety correlates and what they tell us: A literature review. Computers in Human Behavior, 10(3):369{376, Sep 1994. [MRR+10] J. Maloney, M. Resnick, N. Rusk, B. Silverman, and E. Eastmond. The scratch programming language and environment.
Trans. Comput. Educ., 10(4):16:1{16:15,
November 2010. [MT08] [Pap80]
L. Murphy and L. Thomas. Dangers of a xed mindset. ACM SIGCSE Bulletin, 40(3):271, Aug 2008.
Inc., New York, NY, USA, 1980. [Res14] [RF14] [RM+09]
K. J. Reid and D. M. Ferguson. Do design experiences in engineering build a growth mindset in students? In 2014 IEEE Integrated STEM Education Conference, pages 1{5, March 2014.
M. Resnick, J. Maloney, et al. Scratch: Programming for all. Commun. ACM, 52(11):60{67, November 2009. [SON05] A
It included eight Likert-type (1 to 6) items, described in [Dwe00, p. 285]. We used an Italian translation. Questions with * indicate a xed mindset, so they were reverse scored (so that high agreement corresponds with a growth mindset for all questions) Q1* You have a certain amount of intelligence, and you can't really do much to change it.
Q2* Your intelligence is something about you that you can't change very much.
Q3 No matter who you are, you can signi cantly change your intelligence level.
Q4* To be honest, you can't really change how intelligent you are.
Q5 You can always substantially change how intelligent you are.
Q6* You can learn new things, but you can't really change your basic intelligence.
Q7 No matter how much intelligence you have, you can always change it quite a bit.
Q8 You can change even your basic intelligence level considerably.
A.2
Q1 I feel insecure about my ability to interpret a com
puter printout Q2* I look forward to using a computer on my job Q3 I do not think I would be able to learn a computer
programming language Q4* The challenge of learning about computers is ex
citing Q5* I am con dent that I can learn computer skills Q6* Anyone can learn to use a computer is they are
patient and motivated Q7* Learning to operate computers is like learning any new skill, the more you practice, the better you become Q8 I am afraid that if I begin to use computer more, I will become more dependent upon them and lose some of my reasoning skills Q9* I am sure that with time and practice I will be as comfortable working with computers as I am in working by hand Q10* I feel that I will be able to keep up with the ad
vances happening in the computer eld Q11 I would dislike working with machines that are
smarter than I am Q12 I feel apprehensive about using computers Q13 I have di culty in understanding the technical as
pects of computers Q14 It scares me to think that I could cause the computer to destroy a large amount of information by hitting the wrong key Q15 I hesitate to use a computer for fear of making
mistakes that I cannot correct Q16 You have to be a genius to understand all the spe
cial keys contained on most computer terminals Q17* If given the opportunity, I would like to learn more
about and use computers more Q18 I have avoided computers because they are unfamiliar and somewhat intimidating to me It included nineteen Likert-type (1 to 5) items, de- Q19* I feel computers are necessary tools in both eduscribed in [HGK87] (we used an Italian translation of cational and work settings the slight variation proposed in [SON05]). Questions with * indicate a low level of anxiety, so they were reverse scored (so that high agreement corresponds with high anxiety for all questions)