Here’s a typical process for getting into a committed relationship via online dating:

• Swiping through profiles of potential matches
• Initial match based on photos and profile descriptions
• Testing out the connection via text chat
• Dating in person
• Serious relationship
• Marriage/cohabitation

There’s a clear progression here. Each successive step involves fewer potential partners. Also, each step is more intensive, in terms of time, emotional energy, and the amount of information you have about the other person.

Here’s a typical process for getting a job as a teacher in a public education system:

• Applying for teaching position(s)
• Initial eligibility screening based on paper qualifications, etc.
• Testing and/or interviewing
• Appointment to the civil service, perhaps after a nominal probation period

Like the online dating process, each step involves fewer people and more information and time.

But there’s a crucial difference. Unlike online dating, teacher recruitment typically jumps straight from the initial screening and the testing in a semi-artificial environment to the permanent commitment. It skips the messy but crucial ‘trial run through in-person experiences’ phase that dating and serious-but-not-necessarily-long-term relationships offers.

Of course, I’m being slightly unfair to teacher career structures, in that the ‘initial eligibility screening’ step usually means that candidates would have gained admission to and gone through pre-service teacher education even before the ‘applying for teaching positions’ step.

Here’s the problem: many pre-service teacher education programmes focus on abstract, theoretical knowledge of education, instead of the specific content knowledge and practicable pedagogical competencies necessary for cultivating student learning in challenging classroom contexts.^* Personally, I got an ‘A’ in classroom management during my postgraduate diploma in education, but I often failed to maintain discipline in my classrooms of cheerfully unruly teenagers. In short, in many education systems, committing to a long-term career as a teacher because you did well in pre-service teacher education is like deciding to marry someone because they sound like a great person based on the anecdotes that their cousin always talks about during lunch at your office cafeteria.

Which leads to the bigger lesson that online dating can offer to teacher careers:

According to my field informant, a common piece of advice for finding a good match via online dating apps is that you should on several dates with each potential partner rather than making snap judgements from slightly awkward selfies or somewhat unnatural first dates. First impressions and initial interactions can be affected by all kinds of factors—the weather, what they had for breakfast, what their boss said that afternoon, whether their nervous habits come across as cute or irritating—that have little impact on the quality of a long-term relationship.

Similarly, most of the information that can be quickly gleaned about potential teachers has little impact on the quality of their long-term contributions to students’ thriving.

For example, in a study of public school teachers in Pakistan, Natalie Bau and Jishnu Das found that ‘observed teacher characteristics explain no more than 5 percent of the within-district variation in TVA [teacher value-added, i.e. how much a teacher contributed to their students’ test scores]’ (p. 64, emphasis added). This is despite the fact that this study had unusually rich data on observed teacher characteristics. It included not just their training and educational qualifications (which would typically factor into public-service employment screenings) and job-related information (years of experience, contract status, whether the teacher was from the local area), but also their content knowledge (measured by their scores on the same tests that were administered to students).

As Deon Filmer, Vatsal Nahata, and Shwetlena Sabarwal put it in a blog on their study of teacher effectiveness in Tanzania:

Everyone agrees that teachers matter a lot for student learning (see research from the US, Uganda, and Pakistan), but there is no consensus on what precisely it is about the teacher that matters. Not just that, but the variables that drive teacher selection, such as qualifications, experience, test scores, training, and professional development turn out to be weak predictors of teacher contributions to student learning.

And here are some of the findings from their study, which used machine learning to identify the characteristics of teachers who contribute most to the learning outcomes of children in Grades 2 and 3:

For Math, the teacher belief that they can help disadvantaged and struggling students learn; the teacher practice of providing clear and helpful written feedback (on homework and tests); and the teacher preparation in teaching foundational concepts are the three most predictive factors for student learning gains. For Kiswahili, where teacher (and other observable) covariates are on the whole less predictive, teacher preparation, practice, and beliefs still emerge as being important (p. 2, emphasis added).

These crucial aspects of teachers and teaching are just too granular to be captured in the standardised administrative data and civil service assessments that are typically used in teacher selection. In contrast, the Tanzania study data included detailed lesson observations as well as a self-report teacher questionnaire that included a section on beliefs and mindsets.

But even when teacher career structures incorporate microteaching lesson observations and interview questions about beliefs into their selection processes, that only addresses part of the problem, i.e. the limitations of narrow, standardised data. The other part of the problem is that one-off, high-stakes selection data might turn out to be inauthentic.

If you want to marry a person who is genuinely kind-hearted, you should pay attention not only to how they treat the staff on your first date at a nice restaurant, but also—and even more importantly—to how they treat you and others around them during moments of stress and disagreement.

The fundamental value of going beyond one-off, standardised data in teacher selection is fundamental to a set of principles for teacher careers that Lant Pritchett and I propose in a primer synthesising relevant research across country contexts and academic disciplines. We call these principals the 5Cs: Choose and Curate toward Commitment to Capable and Committed teachers.

Given that the 5Cs are concise but potentially confusing, here’s a quick overview of what we mean:

• Children need and deserve teachers who are not only technically capable but also motivationally committed.
• Such teachers deserve the stability of a long-term employment commitment from education authorities and employers.
• Identifying such teachers can be difficult, for all the reasons described above. So education authorities and employers should not only choose teachers using the best available information at a few specific points in time (e.g., entry to pre-service teacher education, entry to the teaching profession, appointment to long-term positions), but also use the novice/probationary period as an ongoing process of curation—by which we mean both extensively supporting novice teachers to develop their craft and iteratively identifying those who are most capable and committed to children’s flourishing.

We call the 5Cs a set of principles rather than a model or a blueprint because they can and should be applied differently in different contexts (see Part 4 of the primer for some examples of what this looks like in practice). With that in mind, here are some examples of what the ‘choose’, ‘curate’, and ‘commitment’ principles can look like—both in teacher careers and in online dating.

To push the analogy just a bit further, another similarity between teacher selection processes and online dating is that their ideal outcome is a relationship that both partners feel enthusiastic about—rather than a relationship where one or both partners feel like they’ve ‘settled’.

We imply this in the 5Cs by saying not only that education authorities and organisations should eventually make a long-term commitment to teachers, but also that teachers should themselves be committed to the shared purpose of cultivating children’s flourishing.

Classroom teaching and learning is inherently relational. Personal interactions between students and teachers are vital not only for tailoring instruction to children’s cognitive needs, but also for supporting their socioemotional growth. And with anything personal or relational, each participant’s intentions matter crucially.

So when it comes to education systems and teachers, we don’t want a situation where one partner has settled for someone who seems more or less good enough because they didn’t have a clear idea of what they wanted or needed in the relationship (e.g., when the Indonesian government’s screening process for teachers defaults to the same basic competence tests and passing grades as for prison guards).

Similarly, we also don’t want situations where:

• one partner has settled for the relationship because they were brutally rejected by their not-so-secret crush (e.g., when teaching is treated as a profession of last resort after medicine, law, and engineering);
• one partner secretly dislikes the other partner but pursued the relationship because they want to share in the other partner’s wealth (e.g., when the only approach to improving teacher careers is raising salaries without strengthening professional norms and other working conditions that shape teacher motivation); or
• one partner is sacrificing their wellbeing to stay with a disrespectful partner for the sake of their kids (e.g., when passionate teachers keep accepting low-paid, short-term contracts with no progression opportunities).

Here’s hoping for better matches, stronger relationships, and greater thriving for children in all our education systems.