Are foreign language learners’ enjoyment and anxiety specific to the teacher? An investigation into the dynamics of learners’ classroom emotions

Two teachers. Same students. Same language. One class feels electric — you're leaning forward, laughing, and scrambling to keep up. The other feels like a countdown to the bell. What actually separates those two experiences? Not a vague impression, not a personality clash you can't name — is there a measurable number that captures it? Jean-Marc Dewaele and Livia Dewaele went looking for one. And what they found is that the difference lives almost entirely in enjoyment, not anxiety. For decades, research in second language acquisition foregrounded fear. Foreign language classroom anxiety — FLCA — was the dominant emotional variable. It captures the nervousness many learners report: heart pounding before being called on, panic when speaking without preparation, and embarrassment at volunteering answers. Horwitz, Horwitz, and Cope established the construct in 1986, and it quickly became the lens through which classroom emotion was studied, linked to a range of learner outcomes and measured in thousands of classrooms. Then, roughly a decade ago, the field shifted. Positive psychology — the empirical study of how people thrive rather than merely cope — arrived in applied linguistics. Dewaele and MacIntyre's 2014 study of one thousand seven hundred forty-six language learners worldwide introduced foreign language enjoyment, or FLE, as a parallel construct. They found that both emotions varied with learner-internal and learner-external factors, and that speaking in front of peers was frequently described as both highly anxiety-provoking and genuinely enjoyable. Subsequent work suggested FLE might be more closely tied to the teacher than FLCA was. But there was a gap nobody had closed: no one had directly compared, at a single point in time, whether the same learners felt different levels of FLE and FLCA depending on which of their two teachers was in front of them. That is exactly what this study does. Forty secondary school students in London — seventeen from Dame Alice Owen's School and twenty-three from Westminster School — all reported having two teachers for the same foreign language. Most studied French, with smaller groups in Spanish and German. These were strong students: self-reported test scores ranged from sixty-nine to one hundred percent, with a mean of eighty-nine point five. The design is unusually clean. Same cohort, same language, same questionnaire — students rated both teachers on identical instruments, making the comparison a within-subject one. The students themselves labelled whichever teacher they saw most often as the Main Teacher, and the other as the Second Teacher. The questionnaire measured FLE with ten positively phrased items on five-point scales — things like "I enjoy it," "It's fun," and "We laugh a lot" — and FLCA with eight items adapted from the standard anxiety scale. Predictor variables captured attitude toward each teacher, how frequently the teacher used the foreign language in class, the proportion of time students spent speaking, and perceived teacher unpredictability. Internal consistency was high: Cronbach's alpha above point nine for all four scales. Both non-parametric and parametric analyses were run and produced similar results, so the authors reported paired t-tests as the primary approach. The headline finding is clean and striking. A paired t-test showed FLE was significantly higher with the Main Teacher: t with thirty-nine degrees of freedom equals three point zero one, a p-value below point zero zero five, and Cohen's d of zero point three six — approaching a small effect. Foreign language classroom anxiety, by contrast, showed no significant difference between the two teachers: t of zero point six one, a p-value of point five five, and Cohen's d of just point zero four. Same students, same language, same anxiety. Different enjoyment. The predictor variables tell you why. Three of the four teacher-related predictors differed significantly between the Main and Second Teacher. Attitude toward the teacher was more positive for the Main Teacher, with a Cohen's d of point five seven — between a small and medium effect. The Main Teacher used the foreign language in class more frequently, with a similar effect size of point five eight. And the Main Teacher was rated as less predictable than the Second Teacher, with a Cohen's d of point four six. The amount of time students spent speaking, however, did not differ significantly between the two teachers. The item-level analysis is where the picture sharpens. Of the ten FLE items, five showed significantly higher scores with the Main Teacher, and all five clustered around the emotional atmosphere of the classroom. "I'm not bored" had the strongest effect size at point five two. "I enjoy it" and "It's fun" followed closely. "It's a positive environment" and "We laugh a lot" completed the set — five items about excitement, humor, and a sense that something good is happening in the room. The five items that did not differ significantly between teachers were qualitatively different: "I'm proud of my accomplishments," "It's cool to know the foreign language," and "I'm a worthy member of the group." These reflect stable personal orientations and longer-term group dynamics — things that don't reset when a different teacher walks in. That contrast is the interpretive core of the paper. FLE behaves like a state — it responds to what the teacher does in the room, moment to moment. FLCA behaves more like a trait — tied to the learner's dispositional anxiety rather than to who is standing at the front of the class. This maps directly onto prior work. Dewaele and MacIntyre found in 2019 that attitude toward the teacher explained twenty-four percent of variance in FLE, while the personality trait emotional stability explained thirty percent of variance in FLCA. The teacher is central to enjoyment; the learner's own dispositional profile is central to anxiety. Earlier work by Dewaele, Witney, Saito, and Dewaele in 2018 found that attitude toward the foreign language and attitude toward the teacher together explained the largest chunks of FLE variance — effect sizes of point two nine and point two seven respectively — while FLCA was most powerfully predicted by how advanced learners were and their relative standing among peers. And Dewaele and colleagues in 2019 showed that teacher characteristics like friendliness explained close to twenty percent of variance in FLE but only about eight percent in FLCA. The pattern across studies is consistent: the teacher moves the enjoyment needle, not the anxiety needle. The pedagogical implication is direct. The paper links teachers' capacity to regulate classroom emotion to trait emotional intelligence and to teacher happiness — citing work by Dewaele, Gkonou, and Mercer, as well as by Moskowitz and Dewaele. If enjoyment is teacher-regulated, then what teachers do, how they're trained, and how classrooms are managed matters substantially for student affect. As Arnold is quoted in the paper: teachers who show interest and empathy, who reduce student anxiety and build student confidence, "will have more positive results with their teaching." That is not a platitude when you have the item-level data to back it up — five specific dimensions of classroom atmosphere that move in lockstep with the teacher. The authors are clear about what the study cannot establish. Forty students is a small sample. Two London schools — both presumably selective, given the high mean test scores — are not representative of language learners broadly. Data were collected at a single point in time, with students reflecting on roughly six months of classes. Full anonymity was required for ethics approval, which meant no teacher names were collected and no finer-grained analysis of individual teacher pairs was possible. Qualitative follow-up was ruled out by the same ethical constraints. Within those limits, the core finding holds. And it points toward a question worth sitting with: if enjoyment is this responsive to teacher-related factors — experience, use of the target language, and the capacity to be unpredictable and alive in the classroom — then how schools assign teachers, what they reward in teacher evaluations, and what they prioritize in teacher training all look like questions with a measurable emotional cost attached to getting them wrong. Anxiety, it turns out, you largely carry in. Enjoyment, you largely walk into. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.