Research in spreadsheet management proved that the overuse of slow thinking, rather than fast thinking, is the primary source of erroneous end-user computing. However, we found that the reality is not that simple. To view end-user computing in its full complexity, we launched a project to investigate end-user education, training, support, activities, and computer problem solving. In this project we also set up the base and mathability-extended typology of computer problem solving approaches, where quantitative values are assigned to the different problem solving methods and activities. In this paper we present the results of our analyses of teaching materials collected in different languages from all over the world and our findings considering the different problem solving approaches, set in the frame of different thinking modes, the characteristics of expert teachers, and the meaning system model of teaching approaches. Based on our research, we argue that the proportions of fast and slow thinking and most importantly their manifestation are responsible for erroneous end-user activities. Applying the five-point mathability scale of computer problem solving, we recognized slow thinking activities on both tails and one fast thinking approach between them. The low mathability slow thinking activities, where surface navigation and language details are focused on, are widely accepted in end-user computing. The high mathability slow thinking problem solving activities, where the utilization of concept based approaches and schema construction take place, is hardly detectable in end-user activities. Instead of building up knowledge which requires slow thinking and then using the tools with fast thinking, end-users use up their slow thinking in aimless wandering in huge programs, making wrong decisions based on their untrained, clueless intuition, and distributing erroneous end-user documents. We also found that the dominance of low mathability slow thinking activities has its roots in the education system and through this we point out that we are in great need of expert teachers and institutions and their widely accepted approaches and methods.
It is mutually agreed that “real” computer problem solving is mostly related to professional programmers, so end-users do not have to carry out such activities, since they are highly supported by “user-friendly”, “full-proof” environments. Two of the most frequent phrases in end-user computing is “I can use it, that’s enough” and “I can learn it on my own”. This is in accordance with Panko’s finding, which claims that end-user computing is “invisible to IT professionals, corporate managers, and information systems (IS) researchers” [
Research found that those involved in end-user computing―central corporate IT groups, general corporate management, information systems researchers [
It is mutually agreed upon that end-users are those participants of the digital world who do not carry out any “serious” computer related activities, which remains the privilege of the trained professional of CS/Informatics (Computer Sciences/Informatics).
As it is mentioned above, end-users are usually untrained or self-trained, and they seem to be invisible to digital professionals. On the other hand, they can be self-confident to the extremes, especially those who are trained in Informatics and find themselves in end-user roles. (The sample of
To clarify end-user activities and roles in the digital world, we launched a project focusing on computer problem solving (CPS). Our project is different in nature from previous studies, where only one aspect of end-user computing is thoroughly examined―spreadsheet, text, presentation, etc. management. We claim that CPS is an umbrella project, since end-user activities are considered in symbiosis, where knowledge transfer plays a crucial role in developing computational thinking.
In the present paper, we argue that (1) end-user computing should not differ in nature from any other problem solving, (2) effective teaching methods and approaches can be adapted from other sciences (3) novel methods can and have to be invented and introduced, (4) what teachers do matters, and (5) what “some” teachers, the expert teachers, do matters [
Considering that we launched a pilot project, our first concern was the data collection and the analyses of the available sources in the subject, in different languages and countries. The sampling process was planned in advance, but during
the work, without any previously published guides, we had to reframe our methods and strategies. (Reasons for this: Lack of financial support, lack of availability, language barriers, the diversity of countries, lack of acceptance of the subject, etc. However, we would like to express our gratitude for all the volunteers for helping us.)
We planned to collect teaching, learning materials developed both for institutionalized and self-trained educational purposes, user guides, helps for supporting daily end-user activities, and sources of widely accepted approaches to end-user computing and teaching. At the time of the publication, we hold 85 and 23 printed materials in our native and foreign languages, respectively. Beyond the printed materials we heavily rely on sources available on the Internet and in the built-in helps. The number of digital sources increases more rapidly than of the printed sources for at least three reasons: (1) their easy access, (2) the online versions of the printed materials, (3) and the changing proportion of the printed and digital sources. Beyond the official supporting sources, we have a huge private collection of end-user documents, the results of end-user activities. This collection consist of several hundreds of documents downloaded from the Internet, sent in emails, collected in our previous projects, and collected by supporting researchers and students.
After collecting, analyzing, and evaluating the content of the sources, the results were compared to the results of recent studies in education, thinking modes, and problem solving. This comparison clearly revealed the discrepancies which led us to end-user bricolage, ineffective end-user computing.
In the present paper, we provide samples of the collected sources (Section 5) and the reason why most of the teaching and supporting materials do not match the requirements of effective end-user problem solving. Those materials which are found supporting are mentioned in Section 7, but their number is infinitely small, compared to the number of the analyzed materials. Here, however, we have to mention that our collection is only a thin slice of the existing materials, and consequently, we are open to any further sources which we do not have access.
To talk about different computer problem solving approaches we all need a typology which consists of all the possible approaches and their definitions. At the beginning of our project, we were faced with the lack of such typology; back then, there was no available complete typology which consisted of all end-user problem solving approaches. Consequently, our major concern was at that time to reveal what other sciences and computer programming had already recognized. We found, by comparing the contents of several typologies, that there is a considerable overlapping, however (1) the terminology used in the typologies and (2) the non-recognition of end-user computing would explain that these similarities are not obvious and that the typologies do not cover all end-user problem solving approaches. The results of our analysis, comparison, and extension are summarized in Section 2.1.
In the following phase of our project, we proposed the extended typology of end-user problem solving approaches, where the mathability levels of the named approaches were set up. Such ranking of the mathability levels allows us to a quantitative measuring system of the recognizable methods (Section 2.2).
Our typology of computer problem solving approaches (published in 2014 in Hungarian) [
Some methods of these approaches are adapted from other sciences, while others are recognizable mainly in computer environments, computer related problem solving, and in data and information management.
The concept based approach (DA class), the algorithmic and the information based approaches (SA class), are well defined in Case & Gunstone’ typology for problem solving in sciences [
For emphasizing that building algorithms (Booth―operational) and applying algorithms (Case & Gunstone―algorithmic based) requires different thinking modes, slow and fast, respectively [
Since our concern was to identify all the end-user computing approaches, we had to expand the already existing typologies and added the Trial-And-Error Wizard based (TAEW) SA methods [
However, TAEW is so widely accepted and practiced in computer related activities that we cannot ignore it anymore. The question is why TAEW and information based approaches are so popular in the digital world, while they seem to be the detours of computer problem solving. For the presence of the TAEW based approach one of the best explanations can be adapted from Polya [
“We know, of course, that it is hard to have a good idea if we have little knowledge of the subject, and impossible to have it if we have no knowledge. Good ideas are based on past experience and formerly acquired knowledge.” ( [
Considering the information based approach, Polya stated that
“Mere remembering is not enough for a good idea, but we cannot have any good idea without recollecting some pertinent facts; materials alone are not enough for constructing a house but we cannot construct a house without collecting the necessary materials.” ( [
Rewording Polya’s note in the digital world, we find that computers alone are not enough for solving problems but we cannot solve problems without “collecting” the necessary hardware and software tools. However, we have to note here that it is difficult to find the border line between the information and the TAEW based approaches, since both focuses on the tools, without considering the problem; “I can use it, that’s enough”. The difference between them is rather on the proportion of learned materials. The ECDL exams, for example, require the candidates to follow a long list of familiar instructions, without giving any thought to the problems. These participants of the digital world mainly carry out information based activities during the ECDL exams. However, in other situations, this knowledge is hardly useful, so TAEW based methods are applied outside of the exams. We also found that most of those end-users who never had the opportunity to participate in any formal digital education primarily carry out TAEW based bricolage and they cause serious financial losses [
“The worst may happen if the student embarks upon computations or constructions without having understood the problem.” ( [
The results of one further research let us complete our typology, namely research in mathability [
“This usage has basically two forms:
(1) In some cases we use existing functions and methods provided by a system, and we apply these tools to solve the problems.
(2) Another possibility is, if we, based on existing means of the system, develop new programs and functions for solving new problems.”
In our typology-building process, we recognized that the two approaches to the usage of computer tools matches the two hypernym classes of problem solving approaches. The first usage covers the surface approach (SA) methods, while the second usage the deep approach (DA) methods. This finding of ours, led us to further consequences: based on our completed computer problem solving typology [
The adaptation of the mathability of software tools to computer problem solving approaches allows us to create a quantitative measuring system [
With the association of the typology of end-user problem solving approaches and the mathability of software tools, we invented the mathability of end-user problem solving approaches, which provides us guide lines in recognizing the mathability of problem solving approaches, in teaching methods, in tasks presented in classes, in teaching materials, in exams, in developing curricula, frame works, etc. With this measuring tool we would be able to provide comparable quantitative values associated to teaching and guiding materials.
Due to the facts that our end-user problem solving typology is an extension of previously published typologies in sciences and computer programming and that problem solving in general in the digital era is mostly computer based, our typology would be accepted in other sciences and programming also. The introduction of our mathability associated typology into other subjects is our further concern.
The arrows in the typology of computer problem solving (
If the problem is too complicated we can simplify it, until a stage is reached which leads to discussable or acceptable results. This second option, namely the acceptance of partial results, is a well-known phenomenon in problem solving. Polya suggested that if the problem seems too difficult
“…we must look around for some other appropriate point of contact, and explore the various aspect of our problem; we have to vary, to transform, to modify the problem. Could you restate the problem?” ( [
What Polya suggested, Kahneman [
“This is the essence of the intuitive heuristics: when faced with a difficult question, we often answer an easier one instead, usually without noticing the substitution. The spontaneous search for an intuitive solution sometimes fails―neither an expert solution nor a heuristic answer comes to mind. In such cases we often find ourselves switching to a slower, more deliberate and effortful form of thinking.” ( [
In general, we can conclude that in our typology we can recognize four approaches which require slow thinking: the DA methods for planning, substituting, and discussing―concept and CAAD based approaches―and the two lowest mathability approaches for the aimless usage of the tools―information and TAEW based. There is only one computer problem solving class which utilizes fast thinking; this is the algorithmic based approach (SA class). The proportion of slow and fast thinking approaches is 4:1. Being aware of this number, there is no wonder that end-user computing is erroneous.
Activities at Level 4 have multiple purposes. Beyond building algorithms and discussing and debugging the outputs, the construction, association, and accommodation of schemata [
These findings are in complete accordance with Kahneman, when he states that System 2 (thinking slow or Attention Thinking Mode, ATM) is the only one that can follow rules, compare objects on several attributes, and make deliberate choices between options, and a reliable System 1 (thinking fast or Automatic Thinking Mode, AUM) requires knowledge built up in learning and practice [
“As you become skilled in a task, its demand for energy diminishes. … the pattern of activity associated with an action changes as skill increases, with fewer brain regions involved.” ( [
Applying this rule, we can transfer knowledge from the highest mathability levels to Level 3, where System 1 can work. For reducing the errors in spreadsheet management a reliable System 1 should be used, suggested Panko in his paper, entitled “The Cognitive Science of Spreadsheet Errors: Why Thinking is Bad” [
Consequently, the algorithmic based approach is different in nature from the information and the TAEW based approaches in the SA class. The algorithmic based level is where System 1 operates. At this level the stored schemata are activated, and decisions are made quickly based on these schemata. On the other hand, the information and the TAEW based levels heavily rely on System 2, which Panko claimed as erroneous approaches to spreadsheet problem solving. Consequently, we can claim that mathability Levels 2 and 1 are pseudo problem solving approaches. At these levels there is no real problem solving, the goal is to perform satisfactorily in tests or exams and/or to achieve any kind of output. The information based approach [
Kahneman explains that for a reliable System 1 we need intuition and real experts who can trust their intuition. (The definition of intuition is from Herbert Simon and cited in Kahneman ( [
“The situation has provided a cue; this cue has given the expert access to information stored in memory, and the information provides the answer. Intuition is nothing more and nothing less than recognition.”
However, it is difficult to define who the real experts are. Even Klein and Kahneman took time to find the sources of their misunderstanding.
“He [Klein] was more willing to trust experts who claim an intuition because, as he told me, true experts know the limits of their knowledge. I argued that there are many pseudo-experts who have no idea that they do not know what they are doing (the illusion of validity), and that as a general proposition subjective confidence is commonly too high and often uninformative. ( [
The illusion of validity is not unknown and has been researched extendedly. Some of the earliest works is the paper of Kruger and Dunning [
“Real knowledge is to know the extent of one’s ignorance.”
Further considering the subject of true experts Kahneman explains that acquiring a skill is fundamental.
“If subjective confidence is not to be trusted, how can we evaluate the probable validity of an intuitive judgment? When do judgments reflect true expertise? When do they display an illusion of validity? The answer comes from the two basic conditions for acquiring a skill: (1) an environment that is sufficiently regular to be predictable (2) an opportunity to learn these regularities through prolonged practice. When both these conditions are satisfied, intuitions are likely to be skilled.” ( [
Considering the different approaches to end-user computing and the education of end-users, our hypotheses are the following:
[H1] End-user teaching materials do not support high mathability end-user computer problem solving.
[H2] Teachers unconditionally accept low mathability end-user computing teaching materials.
[H3] Computer Sciences/Informatics Education research does not consider end-user computing and problem solving as important as programming.
[H4] End-user-program developers and end-user-programs do not support understanding and effective computer problem solving.
In our hypotheses the generally accepted views, the main tracks, the most widely accepted, and institutionalized approaches are in the focus. However, it is also expressed in the present paper that there are teachers, teaching materials, approaches which realized the dangers and consequences of the institutionalized bricolage, and try to introduce their findings.
The question we need to address is what it is that leads to low mathability end- user activities. To find an explanation we have analyzed both the direct human participants in the teaching-learning process―teachers and education policy makers―and the indirect human participants―ICT (Information Communication Technology), IT (Information Technology), CS (Computer Sciences) curricula, teaching materials, and software supports.
It is wildly accepted that end-user computing is boring [
It is obvious that low mathability approaches are not in accordance with those opinions and approaches which claim that there are various tools beyond declared programming languages which would serve the development of algorithmic skills and the computational thinking of students [
The meaning system model of Chen et al. [
End-user activities are mostly taught with a belief in the fixed nature of science―if it is considered science at all. In this framework the various software tools are separated from each other, and emphasis is placed on toolbars, menus, and the programming languages as something unique to each application and their different versions. This approach is well presented by the teaching materials which we have analyzed (Figures 5-17) and also by the Spreadsheet Competency Framework, launched in 2016 [
In end-user computing the steps of problem solving are reduced to one: “carry out” something, use the computer, the software, the Internet. In end-user terminology this is usage. Most of the end-users are convinced that this is what matters. However, in practice it is nothing more than surface navigation [
In end-user text management, Level 2 activities are present when end-users are forced to learn how to apply various font and paragraph formattings to pieces of texts―they learn the tool―, but they do not plan or design the document
according to its content; they do not know whether applying these tools leads to a typographically correct document or not. In
The task in
According to the lesson plan, the teacher gave instructions which the students had to follow: (1) the teacher wrote the text on the board, (2) the students had to type it, (3) and then the lines and pseudo lines had to be formatted according to the “computer cooking” instructions of the teacher and/or the text, without giving any thought of typography and content.
Unfortunately, one of the greatest institutionalized computer cooking is the ECDL exam system, where long lists of demands have to be performed in pre- practiced tasks (
“…they identify computer science with a computer driving license. They think that studying computer science is not a challenge, and that anybody can learn it. Computer science is not considered a scientific discipline but a collection of computer skills.” [
In
In
Further examples of Level 1 activities in text management are when Space, Tabulator, and Enter characters are typed until end-users think that the arrangement of the text is “not too bad”; it would look acceptable or quite well in the printed form (
One further consequence of not using real contents is that we are losing an effective motivating tool. Contemporary students want to acquire knowledge which is directly adaptable to the real world around them. They do not necessarily like purely educational purpose software, in spite of its good quality. Students want immediate links to everyday life [
If teachers believe in the incremental nature of science, this changes their teaching goals, and developing a deep understanding and appreciation of science in the students becomes the focus of the teaching. However, even with this teacher-belief the teaching-learning process can go astray if passing exams is the main
goal in this process, rather than the use of methods to reveal the dynamic nature of science.
In computer problem solving the dynamic nature of CS/Informatics can be revealed through knowledge transfer, which is primarily based on the high mathability approaches to problem solving.
Real world problem solving, presented through authentic contents, is designed to increase motivation and give opportunities for practice, and ultimately to enable the building of concepts. Students who are interested in the contents are more likely to be interested in the problems based on them. For the development of concepts practice, problem solving, and repeated activities are needed. Similar to other sciences,
“A concept requires for its formation a number of experiences which have something in common. … Concepts of a higher order in a hierarchy than those which a person already has cannot be communicated by definition.” [
In the following end-user activities, we have found, in the comparison of teaching materials which aim is to support office document management, repeated contents general ICT knowledge, over tens or hundreds of pages: how to start the program, what are in the window, in the toolbars, in the menus, what the new features are, how to color, how to create a border, how to open a file, how to save a file, how to save a file in older formats, how to make a selection, and how to copy and move, etc. There is no reference to the fact that these are basic ICT skills, which are commonly found in all software tools, and as such, clearly presents the teachers’, publishers’ and software companies’ fixed belief in the nature of science, that the different applications should be taught, handled, and used in seclusion. The connections and the common features of these similar programs are not discovered, the knowledge transfer between them is ruled out, the revelation of knowing from other situations is never reached, and these programs are not only tools, but the only and ultimate purpose of the teaching and guiding process. Consequently, they do not fulfill the requirements of textbooks [
The other feature of low mathability teaching materials is that there are no problems presented at all, no real world problem solving, and no design; only surface navigation and computer cooking. We have also found that if tasks are presented, they are mostly meaningless―with no content at all, or at most a couple of lines of fictitious content. These types of pseudo-contents are extremely boring (
mathability teaching approaches, their knowledge and problem solving methods stay at low mathability levels, which we found in the TAaAS project [
In
The sample of
It is also remarkable that these meaningless tasks are country and language independent; similar “tasks” can be found in different teaching materials, regardless of language and location. The spreadsheet examples of Figures 10-12 are from teaching and tutorial materials of three different places of the world, in three languages, and they are in complete accordance with the ECDL exam samples of
Help features and materials are the specialties of the digital world, representing both a blessing and curse. These materials are mainly written by programmers untrained in psychology and education, using language understandable to professionals in Computer Sciences and programming.
In the following spreadsheet examples (
In the definition of both the IF() (
“argument”, in IF(): “logical test”, “condition”, “any value or expression”, “evaluated”, in MATCH(): “array”, “relative position”, “item”, “specified value”, “specified order”, “logical value”, “reference”).
It also frequently occurs that these definitions are not correct, and furthermore, the published teaching materials repeat the errors, and teachers also accept them without any more thought.
Presented in
We also found in the helps that the examples focus on the details of the language. In the sample of
The Spreadsheet Competency Framework was launched in 2016 at the EuSpRIG Conference. It also consists of the framework specification, where spreadsheet skills are listed [
Levels of proficiency: (1) basic user, (2) general user, (3) creator, and (4) developer.
Groups of skills: (1) Design and best practice, (2) Reviewing and team working, (3) Basic skills, (4) Efficiency of use, (5) Formulas, (6) Formatting, (7) Charting, (8) Protection and errors, (9) Data analysis, (10) Macros and automation, (11) Development and problem solving.
Since the analysis of the framework is beyond the scope of the present paper, only three of the features are mentioned here. The Basic skills consist of only three items, Access and save files, Read and enter data, and Set up printing. Two and a half of these are not spreadsheet skills, but general ICT skills, while reading data is a skill, which would be essential for problem solving. However, problem solving is only the skills of creators and developers. Even such framework does not require any problem solving from end-users.
Creating formulas are also essential in problem solving. In this framework, the focus is on the functions and formulas, not on the complexity of the problems. A list of groups of functions are presented, without mentioning that the formulas and functions can be used at any level, but the problems should match the level of the students. In this framework, we are again faced with the problems mentioned and listed in connection with the teaching materials and teaching approaches to problem solving: the focus is on the tools, not on the problem solving and not on the effective problem solving approaches.
It is obvious that the low mathability methods, wildly accepted, supported, and influenced by teachers, software companies, teaching and help materials, examination forms, etc., are clearly recognizable in the erroneous document handling, in the lack of knowledge regarding transfers between applications, and in computer problem solving in general. It is clear that low mathability computer solving approaches have a negative impact on students. It is not widely accepted that end-users are not the low quality participants of the digital world. They are just not professional programmers, on one hand, on the other hand, they seem to be misjudged, mistreated, and misguided.
Several of low mathability samples are displayed and discussed in the paper. However, a sample of one of the victim’s work is presented in
The example is selected from a lesson plan of a pre-service teacher of Mathematics in the final year of her tertiary studies. The non-printable characters of the document clearly reveal the bricolage, which is obviously a TAEW based approach. However, the footnote added to the formula clearly reveals that she is not aware of the lack of her knowledge: “At the end of the lines the ‘=’ sign cannot be entered because of the Word program.” She even blames the software not allowing the entering of the “=” at the end of the lines, instead of creating the equations with an Equation Editor (
She is a victim of low mathability teaching approaches in various senses: (1) her knowledge in computer problem solving is so low that she cannot use computers effectively, (2) as an in-service teacher, she would not be able to use computer tools for real problem solving and to motive her students, (3) and
finally, based on her note added to her formula, she is not aware of her lack of knowledge.
An example of misleading Microsoft messages is presented in
How different teachers can handle this warning message?
Microsoft and experienced teachers want to protect end-users and students. Both know that if there is no extension, there is no problem. Consequently, Microsoft decided a couple of versions ago that it is better to hide the extensions, and most of the teachers accept this setting. This is a low mathability approach.
However, there are teachers who display the extension intentionally and teach its role and importance. They also provide various examples of file conversions and cases when the extension is changed intentionally. (e.g. a .csv extension is changed to .txt for a more convenient opening in a European Excel or the .exe extension is deleted to attach the file in an email.) With this high mathability approach students would understand the different file types, the connection between them, their conversions, and that extension can be changed without doing any harm to the content of the file. In Chen’s meaning system model these are the teachers with belief in the incremental nature of science and high teaching self-efficacy, the experts.
Another type of Microsoft low mathability example is presented in
However, grading should not be solved with embedded IF() functions, because Excel has solutions which are more suitable for this type of problem (e.g. the INDEX(MATCH()) composite function) [
Another two low mathability aspects of the MS solution are that in the help three four-level composite functions are presented for the three inputs, which contain A, B, C, D, and F and 89, 79, 69, and 59 as constants.
For calculating the grades a more expert solution can be provided with the following settings:
・ Instead of the four-level IF(IF(IF(IF()))), the INDEX(MATCH()) two-level composite functions is created.
・ Instead of the constants, a side table is set up (e.g. D1:E5) which holds the grade limits and the alphabetical grades, arranged in vectors (D1:D5 and E1:E5, respectively).
・ Instead of the three functions for the three inputs, an array formula can be used [
・ The solution is {=INDEX(E1:E5, MATCH(A2:A4, D1:D5))} formula.
One can argue that the HLOOKUP() and VLOOKUP() functions would be faster. However, we claim that there are so many restrictions on these two functions that from the high mathability problem solving and the meaning system model aspects, they do not worth mentioning. They are another low mathability solution.
While Microsoft’s is a low mathability, ours is a high mathability solution. Non-expert teachers can argue that the {INDEX(MATCH())} array formula is difficult, and students, end-users would not understand it.
However, expert teachers can develop students’ understanding of algorithms and building and applying schemata which support their flexible problem solving. The {INDEX(MATCH())} solution presents integrated knowledge, the application of previously learned algorithms, flexible and high-efficacy solution. These are the features which experts teachers have and are able to develop with their students.
In the following, mainly citations of different sources are presented to prove that there are effective, high mathability teaching methods which can be adapted and applied in end-user computing. As it was mentioned in the previous chapters, we are in lack of approaches which modify the proportion of fast and slow thinking, banish low mathability interface navigation, and focus on high mathability problem solving and schema construction, to build up knowledge.
In the Meaning System Model [
In the present paper we focus on teaching approaches and teaching materials for end-user computing and problem solving. Within this framework we have selected from Hattie’s 16 attributes of expert teachers those which are relevant in this context [
Finally, we return to the computer problem solving approaches, to the two hypernym categories. Hattie found in his research that
“We can make a distinction between surface and deep learning. Surface learning is more about the content (knowing the ideas, and doing what is needed to gain a passing grade), and deep learning more about understanding (relating and extending ideas, and an intention to understand and impose meaning). The claim is that experts are more successful at both types of learning, whereas both experienced and expert teachers are similar in terms of surface learning.”
Experts are more aware of when there is a need to activate System 2 than experienced but non-expert teachers. To be able to find the right proportion of System 1 and System 2 requires being an expert in our major, which is pedagogical content knowledge: Hattie argues that
“…content knowledge is necessary for both experienced and expert teachers, and is thus not a key distinguishing feature… We are not underestimating the importance of content knowledge―it must be present―but it is more pedagogical content knowledge that is important: that is, the way knowledge is used in teaching situations.”
Being aware of the characteristics of the deep and surface approach problem solving methods and the attributes of expert teachers, we go one step further and claim that the problem is not with “overuse of ATM thinking”, since this is needed in real world problem solving to build strategic procedures―imple- mented in System 2 (ATM thinking)―, but using ATM thinking when AUM thinking would do better. As we have seen, expert teachers are those who can teach students how to build schemata and when they can be recalled, and how to find the right proportion of deep and surface learning.
One further reason to find the right proportion of thinking fast and thinking slow is explained by Polya [
Why is there a shortage of expert teachers in Informatics? Why can teachers not see that being experienced is not enough; they have to be expert? Kahneman’s sunk-cost fallacy is one explanation [
“The decision to invest additional resources in a losing account, when better investments are available”.
Teachers have invested a lot in being experienced, most of them are self- taught―similar to spreadsheet developers and users [
“All too often a company afflicted by sunk costs drives into the blizzard, throwing good money after bad rather than accepting the humiliation of closing the account of a costly failure.”
We practice in end-user computing, in general, and research focusing on spreadsheet errors proves that in end-user computing pseudo-experts outnumber the true experts [
Some attempts to change end-users’ computational thinking, their approach to problem solving have been made. However, it seems that without institutionalized support―both from education policy and education research―they will remain isolated and forgotten.
It was declared around three decades ago that functional programming is more effective for developing students’ algorithm skills than imperative languages, due to the simplicity of the languages [
A problem solving approach is also present in spreadsheet management in the “Succeeding in Business with Microsoft Excel” series [
Also in relation to spreadsheets we must mention, as a good example, the attempts in France. They teach spreadsheets as part of Mathematics, and as such focus on the fundamental formulas and functions, and this knowledge is tested in the “Brevet des colleges” [
Ben-Ari focused on text management [
The list of good practices is obviously not complete, just like the low mathability examples in the previous section, but this cannot be the goal of the present paper. In general, the overwhelming proportion of low mathability tools and teaching approaches in end-user computing is clearly present.
In this paper we focus on the training and problem solving approaches of the non-professional participants of the digital world, end-users for short. We report the results of our findings based on the collection and analyses of the available end-user teaching, guiding, self-learning, testing materials, along with the different teaching approaches in different languages and countries. We also summarize our base and mathability-extended typology of the computer problem solving approaches (published in 2014 and in 2015, respectively), which is a quantitative measuring system for evaluating educational contents.
Our analyses of end-user materials led us to the following conclusions:
・ End-user educational sources primarily support low mathability slow thinking activities, where surface navigation and usage of software tools are in the focus [H1].
・ There are hardly any traces of high mathability slow thinking activities, where real world problem solving has to be carried out [H1], [H4].
・ Intuition based reliable fast thinking activities are scarcely supported and carried out, due to the lack of true expertise [H2], [H3], [H4].
・ End-user computing lacks expert teachers who have the appropriate attitude to maximize the impact on learning. The majority of the CS/Informatics teachers are novice and/or experienced, who do not possess the necessary pedagogical content knowledge. As such they themselves do not know and consequently they are not able to teach their students when and how to apply fast and slow thinking effectively. These teachers focus on tools, on surface navigation, and accept low mathability and frequently erroneous teaching and guiding materials [H2], [H3].
・ Sunk-cost fallacy further deepens the problem. The experienced participants of end-user training and support invested so much that they do not want to or are not able to give up their achievements, but stick with them [H2].
We also found during our analytical process that researches in CS/Informatics education have already recognized some of the problems in end-user computing. However, they decided that end-user computing should be banished from CS/Informatics, and only pure programming has to be taught, that is the only way for developing the students’ computational thinking [H3].
However, we claim that end-user computing in the digital world is not less important than traditional CS/Informatics studies. We found that the results of recent researches in educational studies would provide guidelines for fundamental changes in end-user education, training, and support.
・ With expert teachers end-user computing can be taught as effectively and efficiently as programming and other majors in CS/Informatics.
・ The path to professionalism in CS/Informatics and to professional problem solving in other sciences starts at end-user computing.
・ End-users outnumber professionals, so they cannot be ignored and/or banished. Well trained end-users would be those participants of the digital world who have great impact on developments.
We have a large number of experienced teachers (and also a large number of unexperienced and novice teachers), but that is not enough. The enormous number of erroneous documents, the wasting of human and computer resources, the low mathability teaching materials focusing on technical and language details, ill-titled and misleading coursebooks and many other teaching materials, as well as the fact that teachers unconditionally accept these circumstances, all outline the necessity of change. The effect of experienced teachers and software developers on end-user computing is obvious: what they do matters. However, we are faced with a very unfortunate situation in that they have a negative effect on end-user problem solving.
We are in great need of expert teachers to develop students’ computational thinking, the skills required for the digital era, effective end-user computing and computer problem solving. We must emphasize that “it is excellence in teachers that make[s] the greatest differences, not just teachers” [
We need teachers who are ready to give up their “experienced” status and are open to novel high mathability approaches. Avoiding the sunk-cost fallacy [
Csernoch, M. (2017) Thinking Fast and Slow in Computer Problem Solving. Journal of Software Engineering and Applications, 10, 11-40. http://dx.doi.org/10.4236/jsea.2017.101002