What's the worth of having a single CS teacher program aimed at teachers with heterogeneous profiles?
WWhat’s the worth of having a single CS teacher programaimed at teachers with heterogeneous profiles?
Hernán CzemerinskiFundación SadoskyInstituto de Ciencias / UNGS Martín ScassoFundación Quantitas Fernando SchapachnikFundación SadoskyFCEyN / UBA
Date: November 10, 2020
Abstract
There is consensus regarding the relevance of including Computer Science (CS) in official school curricula.However, this discipline cannot be taught on a large scale until there are enough trained teachers who caneffectively lead a class. In this article, we discuss the results of a 400-hour teacher training programconducted in Argentina aimed at K-12 teachers with no CS background. The only requirement to signup was to be an in-service teacher, and therefore there were a plethora of different teacher profiles thatattended the courses. Our research aims at understanding whether a single teacher training programcan be effective in teaching CS contents and specific pedagogy to teachers with very heterogeneousprofiles. Also, we investigate what teachers expect to do with the contents they learn. To assess thesequestions anonymous examinations and questionnaires were given and interviews were conducted withcourse attendees. Even though the majority reached the expected minimum bar regarding CS contentsand pedagogy, significant differences appear in their self-perception as regards career opportunities in CSteaching. Our conclusion is that carrying out CS teacher training for a broad spectrum of profiles may beeffective for promoting CS contents. However, if the goal is to boost teachers’ confidence in teaching a CSsubject, then having a program which focuses on a more restricted selection of profiles would be a betterstrategy.
Although many of the challenges faced by the countries that have decided to start teaching ComputerScience (CS) at school are shared, not all of them can be approached in the same way everywhere. Cir-cumstances, educational context and traditions, availability of infrastructure, teaching material and humanresources (among other things) are different in different countries [9]. In this article we focus on one of thesevarying aspects: the training of school teachers who can start bringing CS contents to classrooms.As in many other countries [4, 7, 24], in Argentina there are just a few well-qualified CS teachers ableto teach the subject to a class. On top of that, in most provinces there are no teacher training centers offering1 a r X i v : . [ c s . C Y ] N ov S courses. Facing this panorama, the National Ministry of Science commissioned Sadosky Foundationthe design of a teacher training program whose final goal was to train other-subject in-service teachers sothat they can incorporate CS contents in their schools. There was a public call to national universities forthem to partner with a nearby teacher training institute and, together, design (based on a common base-program set forth by Sadosky Foundation) and teach a two-year 400-hour CS teacher training program. Theintention was to take advantage of the synergy between discipline experts and teacher trainers who betterknow school teachers and classroom dynamics. The base-program mandated which CS content had to bepart of the training sine qua non , and also that the pedagogical approach should be either inquiry-based [17]or problem-based [26]. It also required that, in order to get certified, teachers should be evaluated by trainers.Since the only restriction to sign up for the program was to be an in-service teacher, there were a lotof different teacher profiles that attended the courses. From Technology to Gymnastics, including NaturalSciences, Arts, Maths and Legal Sciences among others. Some teachers had never had contact with CScontents at all. This experience raised a set of questions that we address in our research. Specifically, bygrouping teachers by profile we investigate:
RQ1.
Have they learned the expected content knowledge (CK)?
RQ2.
Have they learned the expected pedagogical content knowledge (PCK)?
RQ3.
What do teachers expect to do with the contents learned?To answer these questions an evaluation was carried out in 2019 to 165 program attendees based oninterviews, questionnaires and an examination based on standardized criteria. The study was restrictedto the 3 (out of 8) universities that at the time of conducting the research presented herein were teachingthe last module of their programs. The results show some positive signs: the vast majority achieved theminimum expected learning outcomes in both CK and PCK. Nevertheless, significant differences appear intheir self-perception regarding career opportunities in CS teaching.The study shows that if it is aimed at generating favorable conditions for the promotion of CS contents(in a context of absence of CS in curricula), it is possible to carry out a teacher training program for a broadspectrum of profiles. However, if the goal is to instruct teachers who can teach a new school subject, focusingon a more restricted selection of profiles is a better strategy.The rest of the paper is organized as follows. In section 2 we describe the context in which theteacher training program was designed and the challenges to overcome to carry it out. Section 3 presentspreviously carried out actions that created conditions so that the program could be planned and implemented.Characteristics of the training program are introduced in Section 4 and how teacher profiles were defined inSection 5. The evaluation instruments used for conducting the research are described in Section 6. Section 7 Universidad Nacional de Rosario had already completed the program and the remaining four were in earlier stages.
Argentina is the seventh-largest country (around 3M km ) composed of 24 provinces, each of whichhas its own Ministry of Education ruling over its own educational system. Nevertheless, some generalcharacteristics are shared. In all provinces, formal education is divided into three main stages: Pre-primary school.
It includes children from 45 days to 5 years old. The law states that the last two yearsare mandatory (4 and 5 year old children). There are approximately 1.8 M students and 742 K teachers.
Primary school.
It is compulsory and constitutes a pedagogical and organizational unit aimed at trainingchildren from 6 to 12 years of age. There are approximately 4.8 M students and 374 K teachers.
Secondary school.
It is a mandatory stage for teenagers and young people who have completed primaryeducation. Secondary school age ranges from 13 to 17 years old. There are approximately 3.8 M studentsand 425 K teachers.The National Ministry of Education defines general guidelines regarding educational policies. After-wards, the policies are discussed in the Federal Council of Education (FCE), which is the office of agreementand coordination of education policies that takes care of ensuring the homogeneity and articulation of theNational Education System. Its president is the National Ministry of Education and it is in turn integrated bythe highest educational authority of each province. Each province is responsible for implementing what isdetermined by de FCE. In addition, each province is also responsible for teacher training, teacher salaries andrelations with teacher unions. An important topic of FCE’s charter is the definition of the Priority LearningCores (PLC). PLCs describe what students are expected to learn at different scholar stages in any schoolof the country, beyond social or territorial particularities. Each province is then free to shape their syllabias desired, as long as the PLCs are covered. For instance, they could add contents or make decisions onhow to deliver them. Sometimes PLCs are for a specific subject such as Math or History, while others arecross-cutting.Although in the 90s most schools had some kind of IT classes, they were mostly about word andspreadsheet processing. Later on, by the beginning of the 2010 decade, the idea of “digital native” [19]convinced most of the educational community (teachers, students, parents and authorities) that teachingcomputing applications was not necessary anymore. If anything, computers had to be used in other subjectsas support tools. Thus, IT hours were progressively removed from the curricula of most provinces. As inmany countries, that perspective changed in recent years. In September 2018, PLCs of Digital Education,Programming and Robotics for primary and secondary school [12] were developed and released. Neverthe-less, in most provinces there is no specific time slot where such contents could be taught. Therefore, most3chool curricula would have to be rescheduled to incorporate time for Computing at the expense of othersubjects.Different programs were implemented in Argentina during the last decade. Among these,
ConectarIgualdad [5] achieved the greatest reach due to its national scope. It started in 2011 and lasted for 5 years.In March 2016, the program had delivered more than six million netbooks to secondary school students andteachers and state-run teacher training institutes throughout the country. The large computer availabilitywas a pivotal point to start thinking of teaching Computer Science at schools. Nevertheless, various reportspointed out low teacher digital literacy, resistance, and many difficulties associated with a massive use ofcomputers [8, 13]. Although computers were present in schools, teachers were not (generally speaking)fluent with them.Despite the fact that it has been reported that only teachers with formal CS background should teachCS in schools [20], just a very small fraction have a university degree in Argentina. Most of them are trainedin teacher training institutes, which are non-university tertiary education establishments. There are 1487teacher training institutes throughout the country , but only 34 offer IT programs. And what is worse, just afew of them include at least one programming course. The training is mostly about teaching students how touse computers for the workplace, or to use them to better learn other topics. Therefore, in most cases theseteachers act as teaching assistants in other-subject courses.Summing up, a first challenge was how to introduce CS topics in a country in which ( i ) there are 24provinces with different education systems; ( ii ) in most of them there is no room for neither IT nor CS; ( iii ) there are no well-qualified teachers; and ( iv ) there is scarcity of institutes where teachers could be trained.In Argentina, there is at least one national public university with a CS career in each province. Thus,there are university-level CS teachers all over the country. Given the lack of CS training offered, they werenatural candidates to participate in school teacher training programs in their provinces. Furthermore, theywould also have to become local consultants: the idiosyncrasy of provincial education systems is such thata local Ministry of Education is more prone to embrace any educational innovation if there are specialists inits own province and does not have to reach out to others located elsewhere.Although universities usually have outreach programs, CS departments are not used to working withschools. The few that do, mostly work with students, not with teachers. There are many reasons for thisuniversity-school disconnection. One is that, as said before, most teachers do not study at universities, but atteacher training institutes. Another reason is that university teachers do not have formal educational training.This means that they do not have theoretical grounds on pedagogy nor general knowledge about how schoolsystems work, apart from their personal experience as once-students.In this context, a second challenge was how to make at least one University of each province stop beingindifferent to school reality. It required CS university teachers and researchers whose subjects and research https://mapa.infd.edu.ar/ The Sadosky Foundation is a public-private institution based in Buenos Aires that depends on theNational Ministry of Science and the two local IT business chambers. It was created in 2011 and its goal is tofavor the articulation between the scientific-technological system and the productive structure in everythingrelated to Information and Communication Technologies in Argentina.Until 2015, the enrollment of young people in the IT field undergraduate programs in Argentina hadkept stagnant for more than 20 years [1]. Meanwhile, the IT industry was generating close to five thousandnew positions annually [15]. This imbalance affected the growth of the industrial sector. Different authorspoint out that the absence of contact with CS that students have during their time in formal schooling canpartially explain the disinterest that exists around related professions [2, 24]. In response to the enormousshortage of qualified people, since its creation the Sadosky Foundation implements different programs aimedat awakening technological vocations in teenagers, including school visits and on-line programming contests.Aside from the promoting actions mentioned above, in 2013 the Program.AR Initiative was born asa new program of Sadosky Foundation. Its main objectives were: (i) to incorporate CS content in formalschooling, both in primary and secondary school; (ii) to generate an Argentine community on CS pedagogy;and (iii) to develop teacher training devices that accompany this process. Different actions were carried outby the Program.AR Initiative before the teacher training program analyzed in this article took place. Mostof them were deployed in collaboration with national universities of different provinces: for each actiona public call was settled and universities willingly postulated themselves to be part of it. Each time, aninternational jury ranked the universities based on presentations and teams submitted for the call. Onceselected, the universities were funded to participate in the program and received specialized training by aProgram.AR team of experts. The training always comprised instructions on the program itself but alsosome part of pedagogical ground. Initially, actions focused on raising awareness about the importance ofteaching computing in school. Then, the effort was put into the development of teacher training plans.In order to crystallize the importance of including CS in school curricula among different stakeholdersof the education system, regional discussion forums were organized during 2014. School principals andsupervisors, educational policy makers, representatives of teacher unions, academics and technical teams ofprovincial ministries of education were invited to participate. The proposal of these forums was to analyzehow to introduce computing education in schools. The forums were held in 5 national public universitiesthat a jury selected according to two criteria: (i) having previous experience in promoting the teaching ofCS; and (ii) having a close relationship with the Ministry of Education of their respective province. One wasfrom Buenos Aires province, one from Córdoba province, one from Corrientes province, one from Mendoza5rovince and the last one from Santa Cruz province. In total, 2700 people participated. While the analysisof the experience showed that regional debates around the subject are positive, it also became clear that aprofound transformation within the Argentine education system that implies changes in curricula is quitedifficult to implement.A second action consisted in the development of a 70-hour teacher training course whose goal wastwo-fold: ( i ) to raise awareness regarding the importance of including CS contents in school; and ( ii ) toprepare teachers with no CS background for carrying out short programming experiences with their students.The course was based on a previous one carried out by the Universidad Nacional de Córdoba [10] and coveredclassic programming topics such as sequencing, loops, conditionals, variables, procedures, parameters andevent handling. It not only introduced topics, but also proposed concrete activities for teaching them in theclassroom. Moreover, it also discussed the pedagogical content it was based on (inquiry-based learning [17]).Thus, the course worked on CK and PCK at the same time [3, 10, 21, 27]. Its content was edited as ane-book [6] under Creative Commons License.The course was delivered to school teachers by national universities that postulated themselves as hoststhrough an open call. Selected university teams -composed of 2 CS faculty and 4 CS teaching assistants-received a 40-hour training from Program.AR team mostly focused on PCK, because CK was not an issuefor them. During the 2015-2019 period this course was given by 16 public national universities and reached1457 teachers belonging to 838 schools of 12 different provinces. Being aware that the training course described in Section 3 was only a starting point, the next stepof the Program.AR Initiative was to outline a two-year teacher training program consisting of 400 hoursof study. The 400-hour long was chosen because it is what the national regulation prescribes to award aSpecialist ISCED-certified degree that grants career credits [23]. Regulation also establishes that this typeof certification can only be granted by teacher training institutes. Yet, as said, just a few offer IT-relatedtraining, and even those that do, do not possess CS-specific faculty.In 2016 there was another public call, but this time targeted to consortia composed of two teams: onefrom a teacher training institute and the other from a university. While the former would bring educationalprofessionals, the latter would provide CS and Educational faculty. Universities and training institutes had topostulate together for implementing a CS program that certifies teachers upon successful completion. Onceselected, these consortia would get paid for three-year work: the first one for developing the curriculum andlectures and the other two for teaching the courses.The call established that consortium teams had to be made up of six people. At least two from theteacher training institute and three from the university. In addition, all CS team members from the universityhad to have a university degree and at least one of them hold a MSc or a PhD. In addition, they had to have6een teaching for at least three years in their respective institutions. Last, among those from the universitythere had to be a graduate in Educational Sciences with a MSc degree or higher.Eight (out of seventeen) consortia were selected based on presentations and teams submitted by aninternational jury appointed for the occasion. There were three from Buenos Aires province, two fromCórdoba province, one from Santa Fe province, one from Entre Rios province and one from Neuquenprovince. Those from the same province are far from each other and, therefore, do not share their areas ofinfluence. In total, 672 teachers signed up for the training programs.Regarding contents, the common base-program set forth by Program.AR established guidelines. Atleast, topics had to cover programming, computer architecture, computer networks, robotics, computersystems and dataset processing. It was also required that the pedagogical approach was either inquiry-basedlearning [17] or problem-based learning [26]. In addition, in each course both CK and PCK had to bedelivered together [3, 10, 21, 27].Not all consortia organized the curriculum the same way. To begin with, each one had to adapt it toprovincial regulations. In addition, the targeted teachers were not necessarily the same in different venues:this had to be agreed between Sadosky Foundation, the consortia and educational authorities of the respectiveprovinces. Also, formats and ways of delivery had to be acknowledged by both partners of each consortium.For instance, the number of courses and the distribution of contents among them varied across venues.However, all training programs were constantly supervised and evaluated by the Program.AR team to ensurethat all relevant CK and PCK were correctly developed and delivered as required.Our study is restricted to the three consortia that were teaching the last module of the program at thetime of conducting our research. Based on the university partner these are the Universidad Nacional deCórdoba (UNC), the Universidad Nacional de Río Cuarto (UNRC) and the Universidad Autónoma de EntreRíos (UAdER).
In this study, the sample is composed of 165 teachers. Seventy-four percent are female, which accountsfor the prevalence of women in teaching work. The average age is 41.8 years, with a standard deviation of7.8. This is close to the national average of 42.3 years reported in the last national census of educationalinstitutions [14]. Thirty-eight percent hold a university degree and the remaining sixty-two percent completednon-university tertiary studies. Ninety-four percent are officially certified as school teachers.Among program attendees, there is a wide range of areas of expertise. In this study we define theprofiles in terms of teacher previous training, which in most cases coincides with the subject they are incharge of. Those teachers who fit more than one profile are considered belonging to the one closer relatedto CS. Below is the categorization we made: 7 omputing teachers.
In-service teachers having computing or programming backgrounds. Half of themwere trained at teacher training institutes and the others at universities. Twelve percent neither have ateaching certification nor an ongoing one. Some are in charge of computing or programming classes; othersare assistants or technological referents of a school.
Technology teachers.
In-service teachers trained in Technology, Communication, or other technical disci-plines (Design, Industry, etc.). Seventy percent were trained at teacher training institutes. Thirty percenthad prior access to CS-related contents in undergraduate training. The rest is divided between those whoattended short courses and those who define themselves as self-taught. Most of them teach at secondaryschools.
Science teachers.
In-service teachers trained in Mathematics, Physics, Chemistry or related. Fifty-sevenpercent studied at teacher training institutes and the remaining 43% at universities. They all have a teachingcertification. In addition, they are all in charge of a subject related to their training at a secondary school.Just under half do not have previous IT-related training experience.
Primary teachers.
In-service teachers trained at primary-stage training institutes. Ninety-five percentteach at primary schools and 5% at pre-primary ones. In addition, 17% completed university undergraduatestudies and 14% postgraduate studies. Thirty-one percent have no previous IT-related training. Most whohave previous training, did it in short courses and define themselves as self-taught.
Others.
A highly heterogeneous group of teachers trained in Social Sciences, Natural Sciences, Litera-ture, Foreign languages, Artistic disciplines, Physical Education, Educational Sciences, Legal Sciences andAccounting Sciences. Fifty-five percent studied at teacher training institutes and the remaining 45% atuniversities. Among the latter, just a few have a postgraduate degree. Twenty percent have no previousIT-related training. The rest is divided between those who attended undergraduate courses and those whoattended short courses and define themselves as self-taught. All have a teaching certification and teach insubjects related to their training.As mentioned in section 4, not all venues targeted the same teacher profiles. Therefore, the proportionof each profile is not the same in each of them. In UNRC the vast majority are primary school teachers; inUAdER a higher proportion corresponds to computing and technology profiles; as for UNC, the profiles ofthe attendees are more heterogeneous. Table 1 shows the distribution of profiles by venue. Primary and pre-primary teachers are trained in specialized centers for these educational stages, which are not the same as those ofother stages. rofile UNRC UAdER UNC
TotalComputing
Technology
Science
Primary
32 (80%) 2 (4.8%) 12 (14.5%) 46 (27.9%)
Others
Total
40 (100%) 42 (100%) 83 (100%) 165 (100%)
Table 1:
Teachers grouped by profile by venue
Given the objective of establishing relationships between teacher profiles and learning outcomes, acorrelational, non-experimental -i.e., with no control group- and cross-sectional -i.e., at a specific point intime- research design was chosen [22]. The choice of a quantitative approach is due to the need of gatheringunbiased data based on a predefined hypothesis that characterizes a broad population.A paper-based standardized examination [11] was used to evaluate both CK and PCK. It was designedas a criterion-referenced test [25], so it allows us to interpret the results in terms of goals and achievements.It consisted of 31 questions: 24 of them were closed-ended questions and the remaining 9 were open-endedones. Additionally, a digital-based survey was given to explore what teachers expect to do with the contentsthey learn. Furthermore, a focus group (ten teachers) and several interviews (7 with teachers and 3 withteacher trainers) were carried out on the three venues to gather more information.For the design of the examination and the processing of results, a set of methodological criteria wasadopted to guarantee the validity and reliability of the chosen evaluation instruments. To begin with, nineindependent CS experts were consulted for validating the construct [16], whose observations allowed us toweigh the assessment questions. Also, the first version of the exam was subjected to a pilot test at UniversidadNacional de Rosario. The answers to open-ended questions were evaluated by using rubrics, which in turnwere validated by three independent correctors.For each evaluated topic, a continuous synthetic index was constructed by scoring each question on thebasis of the analyses provided by the nine independent CS experts. The mathematical properties of the indexwere tested to account for the goodness of the method used in its elaboration: existence and determination,monotony, uniqueness, invariance, homogeneity, transitivity and conceptual exhaustiveness [18]. By usingthe index, three levels of performance were defined establishing cut points in tertiles, taking as a referencethe amplitude scale. Assessments that do not reach the minimum expected performance for teaching CScontents at school were graded as low , those that show elementary competencies as basic , and those thatexceed expectations as high . 9
Results
All data analyzed in this article correspond to teachers who were attending the last course of one of theteacher training programs. The goal is to assess ( i ) to what extent teachers acquired the minimum CK andPCK required to teach CS contents at school according to the criterion established by the Program.AR team(RQ1 and RQ2); and ( ii ) what teachers expect to do with what they learn (RQ3). In this section we presentand analyze our findings. RQ1 is aimed at finding out whether school teachers were able to acquire disciplinary knowledge andconcepts which are considered essential for teaching CS at school. The contents included in the assessmentwere delivered in all venues. Broadly speaking, they can be divided into two categories: ( i ) programmingand ( ii ) computer systems. For evaluating them, different exercises were included in the examination: Programming.
Teachers were asked to read and produce code that includes basic programming constructssuch as loops, conditionals, variables, functions, etc. Block-based languages and pseudocode were used inthese exercises. In some of them, teachers had to identify errors in programs and propose both modificationsto fix them and alternative programs. In others, they had to compare and do a qualitative evaluation ofdifferent programs that solve the same problem. There were also exercises to analyze programs in terms ofreadability, design, efficiency and use of procedures.
Computer Systems.
In this context, computer systems were defined as the combination of hardware,software, users, data and computer networks. Some questions inquired about what makes a computer a com-puter. For instance, teachers had to identify different artefacts that have an embedded computer (such as anautomatic washing machine); or characterize computers as machines that receive data from an input device,process data and produce information that displays through an output device. Other activities presentedproblems that may appear while using a computer or a computer network and asked about possible ways ofhandling them.Results of the evaluation are exposed in Figure 1. A first positive sign is that most teachers achieved theexpected CK minimum bar, independently of their profile. It shows that previous training does not determinethe development of CS skills.Regarding programming, ninety-two percent of teachers achieved the expected performance, of whichforty percent were graded as high . It can also be observed that there are similar results among Computing,Technology and Others (the heterogeneous group not linked to CS) profiles, which reinforces that not havinga previous training is not an impediment for developing programming skills. An interesting finding is that10 igure 1:
Results on Content Knowledge by profile
Science teachers stood out above the rest. This could be associated with their facility in exercising abstractand logical thinking. Also, it can be seen that the performance of Primary teachers was a little lower than thatof the rest. Still, results show that eighty-five percent of them can bring some CS content to the classroom.As regards Computer Systems, results are even better. Overall, ninety-four percent demonstrated basicskills, of which fifty-seven percent had a very good performance. Again, Science teachers were above therest, but this time their performance was closer to that of Sciences and Technology teachers. Once more,Primary teachers results were below the others, but most of them (ninety percent) managed to reach theminimum expected bar and forty-one percent got high qualification.
The second research question seeks to determine if teachers have acquired -or deepened- the PCKdelivered to them. That is, inquiry-based [17] and problem-based learning [26].The evaluation of PCK was organized in three types of exercises. In one of them, teachers were givenseveral unordered blocks corresponding to different moments of a class and they were asked to sequencethem such that the resulting ordering was compatible with the studied didactic criteria. The second type ofexercises addressed the recognition of didactic strategies. For instance, they had to solve CK problems, andafterwards identify some pedagogical elements and the role they play in the way problems were introduced.The third type of questions were about designing elements of a class aimed at putting the proposed pedagogyinto practice. For example, they were asked to come up with a series of questions that were effective forintroducing an exercise through an inquiry-based approach. Results are shown in Figure 2.It is worth noting that PCK results are worse than those of CK. This may sound striking, since a priori the pedagogical content seemed to be more accessible than the disciplinary one, especially for teachers withless CS-related training. However, some teachers stick to the way they have taught throughout their careersand reject a change in this matter. This is consistent with some testimonies collected in the interviews. For11 igure 2:
Results on Pedagogical Content Knowledge by profile instance, a teacher from Concepción del Uruguay commented:“One of the biggest obstacles I encountered was that many times I was asked to modify my way ofteaching. But this is very difficult for me because I have done it differently since I started teaching.It requires a great effort to think of delivering classes with didactic approaches that are different fromthose that I am used to”.In some occasions, this was even more pronounced among teachers with some CS background. Ateacher trainer highlighted the following:“Our idea was that both CS content and pedagogy had the same status. However, it was much easier forthose who had not previously taught programming to adopt the pedagogy we proposed than for those whohad already been teaching computing. The latter were more reluctant to change their teaching practices”.Regardless of differences with CK results, it is remarkable that around eighty-six percent of teachersachieved the expected minimum learning outcomes. Moreover, similar percentages are observed in each ofthe profiles. This goes beyond what was previously noticed: independently of their profile the vast majorityof teachers not only can learn CS contents, but also understand how to teach them . The effectiveness of this teacher training program should not only be measured in terms of knowledgeacquisition. It should also be weighed whether or not (and eventually how) trained teachers bring whatthey learn to their schools. Thus, RQ3 aims to unveil what teachers intend to do with the training programcontent. This issue was addressed with a self-administered online questionnaire.12he first question straightly asked teachers about their self-perception regarding teaching CS content intheir schools: “Do you feel prepared to include content studied in the training program in the subject youare currently in charge of?”
More than 90% answered positively to the question. Of those who answerednegatively, half argued that they had not yet reached the necessary knowledge. The other half expresseddifficulties in articulating the content with the subject they teach.Next question was finer grained for finding out possible (not mutually exclusive) scenarios in whichteachers conceive themselves working with CS-related contents. Options included ( i ) building themselveslittle pieces of software -such as animations, video games or trivia relative to the topics of their subjects- tobring to school so their students play as users; ( ii ) teaching to build small pieces of software -like animationsor simple applications- to enhance what they already teach in their subjects; ( iii ) teaching CS contentsduring a limited period of time, ranging from a week to a term; and ( iv ) being in charge of a CS subject.Results are shown in Figure 3. Figure 3:
Possible scenarios by profile
Most teachers consider using what they have learned as a new resource in their current subjects some-how, including those that had no previous contact with CS contents at all. A high percentage from allprofiles thinks of making his/her students create animations related to his/her subject contents. Moreover,seventy-five percent of Primary teachers consider delivering CS contents for a limited period of time; andfifty percent or more of those of the other profiles would do it too. It can also be observed that, apart fromScience ones, most teachers plan to develop computer programs and bring them to school so their studentsuse them. This points out that they feel comfortable with the idea of developing some kind of software; i.e.,that they have acquired enough CK for programming computers. A teacher trainer said during an interview:“I think most teachers are today in a position to teach CS contents. Obviously, they have to go oncontinuous professional development to deepen what they have learned and also learn what they stilldon’t know. Anyway, they managed to solve problems they had never faced before by using completely13ew tools for them. Many have also been investigating other related topics on their own, without beingasked to do it. If we have achieved this, all was worth it”.Being in charge of a CS subject is considered possible by forty-four percent of respondents, most ofthem belonging to either Computing or Technology profiles. Contrary to what might be assumed -taking intoaccount the results of the exam- only twenty-one percent of Science teachers consider developing a careerin CS teaching. Also, only eighteen percent of Others teachers see themselves doing it. It might also besurprising that even though Primary teachers had the lowest performance, forty-five percent see as possibleleading a CS class.Previous experience with CS topics has a strong impact on teacher confidence: analyzing responsesof the questionnaire, it is recognized that fifty-nine percent of teachers who had previous undergraduate orpostgraduate training in programming or CS would teach a CS subject. This percentage reduces to thirty-ninein those who had only previous experience in short courses or who define themselves as self-taught. In thosewho had no previous trainig at all, only twenty-four percent would do it.This diagnosis is consistent with what teacher trainers observed. In general, all agree that the vastmajority of attendees have managed to develop basic learning for teaching CS topics. Nevertheless, theyhighlight the different level of depth reached by those who had some formal previous training and those whodid not have it. Moreover, they consider that not all are in a position to be in charge of a CS subject. As thisregards a teacher trainer commented:“Most teachers who did not have previous computing formal training cannot qualitatively improve theproposals that we have given them. They replicate what we do. Maybe there are some that could adaptsome modules that we have offered to their needs, but for being in charge of a CS subject they wouldneed much more disciplinary training”.
The training program analyzed in this paper was aimed to prepare teachers for bringing CS contents toschools. The results give account that the goal was achieved: regardless of their previous training the vastmajority have acquired CK and PCK and consider incorporating CS contents in different scenarios. However,these encouraging results should be viewed with caution when considering the goal of incorporating a CSsubject into school curricula. Teachers’ previous experience, the educational level in which they teach andthe way they go through a training process seem to strongly influence their expectations. Even though it was atwo-year long training program, those who mostly consider themselves capable (or interested in) developinga professional career in CS scholar education are those who already had some previous training in the field.A training program targeted to a wide spectrum of teachers can be very effective if it is aimed at raising14wareness of the importance of including CS contents at school, generating interest in teachers and studentsand including the issue in the educational policy agenda. This is important in the Argentine context, wherethe presence of the theme is only incipient. Nevertheless, if the matter of teaching CS in schools were at amore advanced stage in the educational and political community or if it was intended to generate systematicteaching experiences, it would be more effective to develop a more in-depth training targeted to teachers withprofiles oriented to technical teaching.
References [1] Alonso, M. L., Molina, G., Ramagnano, G., Rico, M. J., Sampaolesi, E., Saulo, H., and Rivera, L. (2016). Síntesis deinformación estadística universitaria argentina 2014-2015, ministerio de eduación de la nación argentina.[2] Bell, T., Andreae, P., and Robins, A. (2014). A case study of the introduction of computer science in nz schools.
ACM Trans.Comput. Educ. , 14(2).[3] Bosse, Y., Redmiles, D., and Gerosa, M. A. (2019). Pedagogical content for professors of introductory programming courses.In
Proc. of the 2019 ACM Conference on Innovation and Technology in Computer Science Education , pages 429–435. ACM.[4] Brown, N. C. C., Sentance, S., Crick, T., and Humphreys, S. (2014). Restart: The resurgence of computer science in uk schools.
ACM Trans. Comput. Educ. , 14(2).[5] Consejo Federal de Educación (2010). Las políticas de inclusión digital educativa el programa conectar igualdad.[6] Factorovich, P. and Sawady O’Connor, F. (2015).
Actividades para Aprender a Program.AR . Fundación Sadosky, 2 edition.[7] Falkner, K., Vivian, R., and Williams, S.-A. (2018). An ecosystem approach to teacher professional development withincomputer science.
Computer Science Education , 28(4):303–344.[8] Fontdevila, P. A. (2011). Estudio de caso: Conectar igualdad.
Revista Iberoamericana de Ciencia, Tecnología y Sociedad -CTS .[9] Hubwieser, P., Armoni, M., Brinda, T., Dagiene, V., Diethelm, I., Giannakos, M. N., Knobelsdorf, M., Magenheim, J.,Mittermeir, R., and Schubert, S. (2011). Computer science/informatics in secondary education. In
Proc. of the 16th AnnualConference Reports on Innovation and Technology in Computer Science Education - Working Group Reports , pages 19–38.ACM.[10] Martinez, M. C., Gomez, M. J., Moresi, M., and Benotti, L. (2016). Lessons learned on computer science teachers professionaldevelopment. In
Proc. of the 2016 ACM Conference on Innovation and Technology in Computer Science Education , pages 77–82.ACM.[11] Martínez Rizo, F. (2001). Evaluación educativa y pruebas estandarizadas. elementos para enriquecer el debate.
Revista de laeducación superior , 4(120):71–85.[12] Ministerio de Educación, Cultura, Ciencia y Tecnología (2019). Núcleos de aprendizajes prioritarios. educación digital,programación y robótica.[13] Ministerio de Educación de la Nación Argentina (2011). Evolución y seguimiento programa conectar igualdad. informe deavance de resultados 2010.[14] Ministerio de Educación y Deportes (2016). Censo nacional del personal de establecimientos educativos.[15] Observatorio Permanente de la Industria del Software y Servicios Informáticos (2016). Reporte anual sobre el sector desoftware y servicios informáticos de la república argentina, reporte año 2015.[16] Padua, J. (1979).
Técnicas de investigación aplicadas a las ciencias sociales . Fondo de Cultura Económica: Sección de obrasde sociología. Fondo de Cultura Económica.
17] Pedaste, M., Maeots, M., Siiman, L. A., de Jong, T., van Riesen, S. A., Kamp, E. T., Manoli, C. C., Zacharia, Z. C., andTsourlidaki, E. (2015). Phases of inquiry-based learning: Definitions and the inquiry cycle.
Educational Research Review ,14:47–61.[18] Pena-Trapero, B. (2009). La medición del bienestar social: una revisión crítica.
Estudios de Economía aplicada , 27(2):299–324.[19] Radford, M. L., Connaway, L. S., Agosto, D. E., Cooper, L. Z., Reuter, K., and Zhou, N. (2007). Behaviors and preferencesof digital natives: Informing a research agenda. In
Joining Research and Practice: Social Computing and Information Science- Proc. of the 70th ASIS&T Annual Meeting, ASIST 2007, Milwaukee, WI, USA, October 19-24, 2007 , volume 44, pages 1–15.Wiley.[20] Ragonis, N., Hazzan, O., and Gal-Ezer, J. (2010). A survey of computer science teacher preparation programs in israel tellsus: computer science deserves a designated high school teacher preparation! pages 401–405.[21] Redish, E. (2000). Discipline-based education and education research.
Journal of Applied Developmental Psychology - JAPPLIED DEV PSYCHOLOGY , 21:85–96.[22] Sautu, R., Freidin, B., Najmias, C., Otamendi, M. A., Paredes, D., Ballesteros, M., Perea, C., Plotno, G., Lederman, F.,Ballesteros, M., et al. (2014).
Metodologia de la Investigacion I . México DF: Interamericana editores.[23] Sentance, S. and Csizmadia, A. (2017). Professional recognition matters: Certification for in-service computer scienceteachers. In
Proc. of the 2017 ACM SIGCSE Technical Symposium on Computer Science Education , pages 537–542. ACM.[24] Shein, E. (2019). The cs teacher shortage.
Commun. ACM , 62(10):17–18.[25] Weiss, D. J. and Davison, M. L. (1981). Test theory and methods.
Annual Review of Psychology , 32(1):629–658.[26] Wood, D. F. (2003). Problem based learning.
BMJ , 326(7384):328–330.[27] Yadav, A. and Korb, J. (2012). Learning to teach computer science: The need for a methods course.
Communications of theACM , 55:31–33., 55:31–33.