The relevance of computer self-efficacy to digital skill
[if you are more interested in the practice than the theory after watching the short video introduction, just go to “Part 2 — practice: Teaching Digital Skills to low CSE students — some design principles]
Part 1 — Theory
Let’s start with some definitions. What is self-efficacy (and for the colleagues who laughed at me when I used the term in the context of digital skills, how do you pronounce it Listen here)?
Bandura (1982) originally defined it as: ‘Self-efficacy is the belief in one’s capabilities to organise and execute the courses of action required to manage prospective situations.’ He pointed out that even though self-belief did not ensure success, ‘self-disbelief assuredly spawns failure.’ Bandura’s social cognitive theory has been written about extensively and used in many domains. Self-efficacy is not about skill level, but ‘reflects what individuals believe they can do with the skills they possess’ (Eastin & LaRose, 2000)
I have often observed that the obstacles students encounter can be less to do with learning how to use a computer and more to do with the beliefs they carry into the learning situation. It may just be possible that student dependency on IT support, is driven by a lack of belief that they can become more self-directed with regard to computer use. Do they lack digital self-efficacy? Can we learn from the way this framework has been used in other domains, and design digital learning interventions that build these self-efficacy beliefs? This is the context for this article.
Computer Self Efficacy — A new concept?
It turns out that although I thought I had coined the term ‘digital self-efficacy’ as something that digital learning courses needed to address, the area exists as a domain of research in its own right: Internet Self-efficacy studies, ICT self-efficacy scales, computer self-efficacy and the impact it may or may not have on academic performance, are all of interest to researchers who have brought together the construct of self-efficacy and computer skill development. Can I build into a Digital Learning Syllabus interventions, possibly early on in the student learning journey, that may develop a degree of self efficacy? Would this make it easier for students to be more self-directed in this domain?
This area of research has been active for quite a few years, yet learning technology as a field seems mostly focussed on skill development as a way to develop confidence in use. Yet, the research does suggest that, it may just be that practice does not make perfect without the belief that we can act efficaciously when we approach a computer.
Compeau and Higgins (1995) in a study carried out in an organisational setting, rather than an educational one, found that ‘computer self-efficacy […] exert[s] a significant influence on individuals’ expectations of the outcomes of using computers, their emotional reactions to computers (affect and anxiety), as well as their actual computer use.’ It seems almost common sense to postulate that if your computer self-efficacy is low, this will affect the cognitive functions needed for effective learning. The authors go further than the self-eviden, however, and propose that understanding computer self-efficacy ‘is important to the successful implementation of systems’.
Could it be that we are so focussed on which shiny new tool to bring in to help students ‘embrace the digital transformation of our campuses’ that we are missing a crucial part of what will bring increased success in the adoption of digital tools in education?
Computer self efficacy (CSE in what follows) ‘is positively correlated with an individual’s willingness to choose and participate in computer-related activities, expectations of success in such activities, and persistence or effective coping behaviours when faced with computer-related difficulties’ (Karsten and Roth, 1998).
“The relationship between self-efficacy and personal computer use is perhaps intuitively obvious. Personal computers represent a complex and somewhat troublesome technology, requiring considerable skill and extensive training to operate successfully. Self-efficacy is essential to overcome the fear many novice users experience.” (Eastin and LaRose, 2000)
I could not have put it better myself!
Enhancing a digital learning syllabus so that it addresses low self efficacy, or that sinking feeling many experience when told ‘it’s easy, you just need to…’ seems a no brainer given these research findings.
Why is this important?
The willingness to experiment is core to one’s ability to use a computer. If a student does not believe they can carry out actions to get the computer to do what they want it to do, how do they ever get started on the road of self-direction? Furthermore, experimentation or ‘futzing around’ as internet slang would have it, is an integral part of the mindset of skilled computer users. We expect tech to fail, we are not surprised when it does, and futz around as needed to get it working again.
Research has found that those with high levels of CSE perform and cope better in remote working situations than those with low CSE. Some studies distinguish between CSE and Internet self efficacy (ISE) and offer the daunting proposal that it requires CSE plus further advanced skills: “these include establishing and maintaining a stable Internet connection, learning how to navigate the Internet, and searching it for relevant [and reliable] information.” (Eastin and LaRose, 2000).
Further, there is a suggestion in the literature that prior experience with related fields such as maths or engineering may offer advantages to novice users in learning speed, and that actual use is what builds self efficacy — some suggesting it needs at least 2 years consistent experience to develop a degree of CSE — many scales exist in the literature that attempt to measure degrees of CSE, but this is not my concern here. Other researchers disagree with the notion that ‘consistent experience’ suffices, and suggest that CSE is something that has to be trained separately to skill or at the very least has to be built in the pedagogical approach of any training.
The necessary antecedents to self efficacy
That self efficacy affects behaviour is widely accepted in this research; there is much less literature on what specifically develops self efficacy in computer use (Marakas, et al. 1998). Bandura postulated 4 sources of development: enactive mastery, vicarious experience, verbal persuasion, and affective states. Enactive mastery has been found to be the strongest source of efficacy information, and it is developed through prior performance of a task — logging into a learning management system; vicarious learning is a source of self -efficacy through observation and demonstration of a task — following a tutorial about an application feature; verbal persuasion is a source through verbal cues such as performance feedback — you can do this, here are the steps to follow; affective states are the weakest source — such as computer anxiety as a negative source that reduces CSE.
In sum, “Self-efficacy is what students infer from the information from these sources […], the judgement they make about their ability to succeed on a specific task or set of related tasks” (Margolis and Mccabe, 2006).
Computer Self Efficacy in Learning Management Systems
Saadé and Kira (2009) used the learning management system as the target technology to further explore the link between CSE, Computer Anxiety (CA) and Perceived Ease Of Use (PEOU). The relationship is complex and, some argue, the parameters that affect each may not be independently defined.
Computer Anxiety can be defined “as a feeling of being fearful or apprehensive when using or considering the use of a computer” (Leso and Peck,1992) and this study found that computer anxiety scores remained unchanged after completion of two 1-year introductory computer science courses — a programming and a non-programming one. Read that again. Anxiety scores remained unchanged in spite of one full year of practice.
Perceived ease of use can be understood as the extent to which the student expects a low level of cognitive effort when using a computer or an application.
Computer anxiety reduces levels of CSE and affects perceived ease of use. These 3 constructs interact in ways not yet clear in research findings. What is clear is that they do affect performance, and have been shown to predict performance more accurately than past computer experience or performance on courses that require digital skill.
Saadé and Kira’s study offered some support for the idea “that the quality of support, friendliness, and enthusiasm to the student using the LMS has a profound influence on the prevalence of computer anxiety”. If anxiety is reduced, CSE increases and perceived ease of use (PEOU) improves thus leading to better performance when using the LMS. The relationship also works via CSE, high CSE decreases anxiety and this improves PEOU. Whilst the constructs interact in complex ways, their study does support the idea that offering a syllabus that tests and intervenes early on at the level of computer anxiety and CSE may lead to better performance in digital skill training.
to better understand the dynamics of human computer interactions, qualitative feelings a person ascribes to some previous computer experience need to be understood. This construct might dominate as a predictor of successful and satisfied usage. Course designers and managers that desire to successfully implement LMSs in a higher education or training context need to be aware of this relationship in order to create an environment supportive of subjective attitude toward the artefact. (Saadé and Kira, 2009)
In other words, producing impeccable resources for students that offer vicarious unsupervised learning can act to either reduce or increase CSE depending on anxiety levels. Cognitive presence in digital skill development seems key in creating a supportive environment for the ‘subjective attitude towards the artefact’ students have.
Telling students they lack skill, that the instructions are clear or easy when this is not their experience, is unlikely to lead to motivation to learn. Yet, this is what students hear far too often from the people they seek help from. It takes much less time to throw out a fish, than to teach the student how to fish. This unconscious process is how educational systems sustain dependent students who are often unable to make use of the extensive resources they are given. We attribute this to lack of digital skill, but it may be something else entirely.
I am now supporting learning technology live coaching with students at my institution, as this enables me to address the skill element alongside the anxiety and the beliefs they bring into the learning situation. In interaction, verbal persuasion techniques can teach the student that anxiety free computer learning is possible. Instead, we often foreground shiny new tools via static resource production, and background important (and needed) cognitive development.
Research has consistently found that CSE is low when students lack a live presence in their journey with an over-reliance on standardised materials that by definition cannot adapt in dialogue to the subjective experience of the student. And no, I do not believe that eye tracking can be a substitute for cognitive presence. Or as my line manager puts it: “Just having someone to bounce a quick question off easily can help so much”. It can also build self efficacy for the future and reduce anxiety. This, in turn, can help develop students who can become their own tech support over the course of their studies and rely on scarce institutional resources less.
This research also speaks to the dearth of good design in the LMS as a learning tool; it is far too often designed for content archiving, surveillance, and data collection more so than for ease of learning. Computer self efficacy (CSE) using the LMS can be increased if the LMS has a user interface design that is easy to use — invisible in task execution and producing a low cognitive load.
When I see the ‘contortions’ students have to go through within the environment of most current LMSs to do the simplest of tasks, I see a negative self-perpetuating loop: each time the student has a negative ease of use experience, accompanied by incomprehensibly long videos on how to upload a video, for example — something they can do in Tik Tok in seconds — their CSE reduces a little more and their anxiety in relation to that tool increases. The belief that they can positively take action in that setting can be slowly eroded as they develop learned helplessness over time; trading agency and the belief that they can use a computer effectively for an over-reliance on tech support tickets. This loop can be reinforced when IT takes the problem away in the moment, and ‘just fixes it’ to manage their own scarce resources.
Yet, CSE research indicates that those with low CSE benefit more from the kind of vicarious learning that involves ‘coping strategies’ — let’s learn together how to fix this problem, than from ‘expert strategies’ — here is an expert tutorial on how to do this (Margolis and Mccabe, 2006). This perhaps further suggests that those of us who offer technical support to students may need more resources and training ourselves to enable the possibility of a learning situation that encourages greater self efficacy.
As I seek an answer to my initial question, I realise the question has changed slightly. Of course I can build CSE development early in any digital learning syllabus. A more pressing question now seems to be: Can I list design principles for any learning technology intervention that will, in process if not content, embed a pedagogy of self efficacy rather than one of expert instruction? I think I can make a start on this helped by Margolis and Mccabe’s work on learning strategies to increase self efficacy.
“Low self-efficacy beliefs, unfortunately, impede academic achievement and, in the long run, create self-fulfilling prophecies of failure and learned helplessness that can devastate psychological well-being.” (Margolis and Mccabe, 2006)
Part 2 — Practice
“…the presence of high support may in some ways actually hinder the formation of high self-efficacy judgments. If individuals can always call someone to help them when they encounter difficulties, they may never be forced to sort things out for themselves and thus may continue to believe themselves incapable of doing so.” (Compeau & Higgins, 1995)
TL;DR for this section — Here is what not to do: don’t keep telling students the instructions are easy when they tell you they do not understand them; don’t ignore students when they tell you they feel anxious using a new device and just keep telling them that there is no need to be anxious; don’t keep fixing things for them to save time; don’t show off how quickly you can do it with a near perfect demo.
Teaching Digital Skills to low CSE students — some design principles
Let’s look at what we can do.
Use tasks tailored to student skill level
When students are learning digital skills, task difficulty has to be set according to what matches their current level of skill in order to increase self efficacy. If there is a mismatch there will be problems. Too hard, it can create anxiety, too easy (or done for them by the IT expert) and it can reinforce low self-efficacy beliefs — I cannot take meaningful action on this device, I cannot be trusted to do this myself.
Use peer models — sitting be ‘Nellie’ training works
Students learn well from each other. In digital learning, students learn from leaning over to another student’s computer and informally checking how something is done; a peer shows them, they get on with studying. This is challenging with remote learning, as often students are alone in a room at home trying to do an assignment when they may want to learn how to get a word processor to create a table of content, for example. In digital skill learning the lack of informal peer learning, could make the difference between passing or failing a course.
Teach learning strategies not linear instructions
Show students the steps to follow in order to find an answer/solution for themselves. Linear instructions that require no thought or learning, but just clicking do not help CSE development. When familiar software is updated, students get lost with yesterday’s intructions and feel helpless because they do have not learnt how to help themselves — let’s check the app’s help and see how we can do this.
To enhance their digital self efficacy students need low anxiety towards the idea of experimenting with their devices. There is an issue of time here, it takes longer to work this way — both for the teacher and the student. It may also require more skilled staff to work this way. There is an issue of the balance between digital security and experimentation also, as often finding your own answers in this context means searching online.
What are their interests?
Decontextualized tutorials are boring. Transferring learning from one context to another is notoriously challenging. If we can find ways to tie our learning material or concepts to the actual interests of the student — say sports, pop culture, or movies — we will help them learn and apply that learning directly to the context where they need it. I have been helping students of Sport Science lately. They look utterly bored as they ask for help with computer use. As soon as I get them talking of how they might use tech in their sport coaching or their own sport, the nature of the conversation changes completely — they can see how technology relates to what they are actually interested in and how it may help them.
Let them choose how to learn
I now create custom web pages for each student with curated resources after our coaching sessions. I tell them that the page will have options and that they can choose what serves and ditch the rest. I also offer different kinds of follow ups: read and post a question, watch a video, try it with me, find a peer for co-learning. My hope is that this supports the idea that offering choice enhances CSE. I will be evaluating this idea over time. Here there is also an issue of time and resource, creating a resource that fits all is much quicker and easier to produce than a resource that is custom and unique to each student. So I use open online curating platforms that can produce a custom page with little effort. Why? Because I believe that offering choice based on well assessed needs, we ultimately save in both time and resource as the student population increases the degree of CSE they bring to their studies over time. A long term educational goal!
Encourage students with formative feedback
Offer clear strategies for students to try to find the answer themselves. Give students consistent, credible and specific feedback. Tell them: You can learn how to do this, here are things you can try, try this feature for 1 day, then review how it works for you. In trying and having to think about what works and what does not, they learn that they can themselves tackle the computer in productive ways.
Encourage accurate attributions
This is a strategy from cognitive psychology which can break down into very complex interventions, but can be introduced to students easily in the way we talk to them. When humans experience anxiety they tend to attribute the anxiety to the wrong stimulus: I fail because I am stupid or because I just cannot understand computers. This is the wrong attribution in a learning context.
It is far more helpful to remind students that if they fail it is because instructions were not followed — we can review those and try again. Or because they did not allow enough time on task — we can do it again together. Or they did not implement the learning strategy suggested — we can review it and walk through it again. This process needs the student to be patient and motivated to learn about the devices they use, this does not always obtain when students are pushed for time and assignments deadlines.
As I implement the ideas explored in this article in my work with students, I have found them useful to reflect on my own understanding of the obstacles they face when using computers for their studies. I do now work more on the assumption that it may just be less about skill development, and more about reframing the beliefs users bring into the digital learning situation.
And may be after all this reading it is worth enjoying the 2 minute summary again:
Acknowlegments
With gratitude to the students that have been willing to share their challenges in our coaching sessions this year, my line manager who is supporting a bottom up inquiry into the methodology best suited for teaching our unique student cohort, and to Jenny Mackness for giving up her free time to make this article more substantial and readable with her excellent editing skills.
Bibliography
Bandura, A. (1982). Self-efficacy mechanisms in human agency. American Psychologist, 37, 122–147. Link pdf
Compeau, D., & Higgins, C. (1995). Computer self-efficacy: Development of a measure and initial test. MIS Quarterly. 19, 189–211. Link
Eastin, M. S., & LaRose, R. (2000). Internet self-efficacy and the psychology of the digital divide. Journal of computer-mediated communication, 6(1), JCMC611.
Karsten, R., & Roth, R. M. (1998). Computer self-efficacy: A practical indicator of student computer competency in introductory IS courses. Informing Science, 1(3), 61–68.
Leso, T., & Peck, K. L. (1992). Computer Anxiety and Different Types of Computer Courses. Journal of Educational Computing Research, 8(4), 469–478. https://doi.org/10.2190/Q1TJ-8JCU-LDAP-84H8
Marakas, G. M., Yi, M. Y., & Johnson, R. D. (1998). The multilevel and multifaceted character of computer self-efficacy: Toward clarification of the construct and an integrative framework for research. Information systems research, 9(2), 126–163. Link
Margolis, H., & Mccabe, P. P. (2006). Improving Self-Efficacy and Motivation: What to Do, What to Say. Intervention in School and Clinic, 41(4), 218–227. https://doi.org/10.1177/10534512060410040401
Saadé, R. G., & Kira, D. (2009). Computer anxiety in e-learning: The effect of computer self-efficacy. Journal of Information Technology Education: Research, 8(1), 177–191.
S. A. Templin, M. Shiroku, & K.Taira (1999). Self-Efficacy Syllabus
The Language Teacher — Issue 23.4; April 1999
Watts, R. E. and Bandura, A. (1996). Self-Efficacy in Changing Societies. Journal of Cognitive Psychotherapy, 10(4), 313.
