Foreign Language Enjoyment and Anxiety: The effect of teacher
and learner variables1
Birkbeck College, University of London
Westminster School, London
Birkbeck College, University of London
Worcester College, University of Oxford
Positive psychology has boosted interest in the positive as well as the negative
emotions that Foreign Language learners experience. The present study examines
whether -and to what extent- foreign language enjoyment (FLE) and FL classroom
anxiety (FLCA) are linked to a range of learner-internal variables and
teacher/classroom-specific variables within one specific educational context.
Participants were 189 British high school students learning various FLs. Levels of
FLE were linked to higher scores on attitudes towards the FL, the FL teacher, FL use
in class, proportion of time spent on speaking, relative standing and stage of
development. Lower levels FLCA were linked to higher scores on attitudes towards
the FL, relative standing and stage of development. FLCA thus seems less related to
teacher and teacher practices than FLE. The pedagogical implication is that teachers
should strive to boost FLE rather than worry too much about students’ FLCA.
Dörnyei and Ryan (2015) pointed out that despite the fact that emotions play a crucial
part in our lives, they have been largely “shunned” by Second Language Acquisition
(SLA) scholars (p. 9). The authors attribute this to the cognitivist tradition in the field
and argue that it is time to overcome the general “emotional deficit” in SLA research.
They wonder how as researchers we can “accommodate positive emotions more
effectively into our descriptions of learner psychology?” (p. 205). This statement
recognizes that the role of positive emotion, although vaguely recognized in the field,
still has a long way to traverse before positive emotion assumes the place it deserves
(Dulay & Burt, 1977, Gardner, 1985; Krashen, 1982; Schumann, 1978) but it is true
that it seems to have remained a little bit in the shadows of the vibrant research into
negative emotions, mostly foreign language anxiety. The situation may be changing
because of the influence of Positive Psychology, the empirical study of how people
thrive and flourish. Positive Psychology wants to broaden the general perspective in
general psychology with its focus on abnormalities, disorders, and mental illness and
the development of ways to reduce pain and learn to cope with negative experiences,
1 Pre-print version of the paper (Article first published online: February 17, 2017)
Language Teaching Research https://doi.org/10.1177/1362168817692161
in favour of the development of tools to build positive emotions, foster greater
engagement, and boost the appreciation of meaning in life and its activities
(MacIntyre & Mercer, 2014). Just as the interlanguage paradigm superseded the error
analysis tradition in the 1970s, with a move away from an exclusive focus on second
language learners’ deficits, the Positive Psychology approach advocates a more
holistic view on humans, which in SLA terms means moving away from the
overwhelming focus on negative emotions (foreign language classroom anxiety -
FLCA) to include L2 learners’ positive emotions, such as Foreign Language
Enjoyment (FLE) (Dewaele & MacIntyre, 2014, 2016; Dewaele, MacIntyre,
Boudreau, & Dewaele, 2016).
As MacIntyre and Mercer (2014) put it eloquently:
Many language educators are aware of the importance of improving individual
learners’ experiences of language learning by helping them to develop and
maintain their motivation, perseverance, and resiliency, as well as positive
emotions necessary for the long-term undertaking of learning a foreign language.
In addition, teachers also widely recognise the vital role played by positive
classroom dynamics amongst learners and teachers, especially in settings in which
communication and personally meaningful interactions are foregrounded (p. 156).
The present study proposes to investigate the role that foreign language teachers play
in orchestrating the emotions of their students, in addition to learner-internal sources
of emotions (Dewaele, 2009).
II Literature review
Research on affect and emotions - mainly negative ones such as language anxiety- has
been vibrant in SLA research since the 1970s. The first studies into the effects of
anxiety on SLA (Chastain, 1976; Kleinmann, 1977) gave contradictory results which
Scovel (1978) attributed to the fact that: “anxiety itself is neither a simple nor well-
understood psychological construct and that it is perhaps premature to attempt to
relate it to the global and comprehensive task of language acquisition” (p. 132).
Looking back at the early research, MacIntyre (2017) agreed with Scovel, explaining
that “not all types of anxiety that can be defined and measured are likely to be related
to language learning” (p. 12).
MacIntyre (2017) argued that the so-called early “Confounded Approach” ended
with Horwitz, Horwitz and Cope’s (1986) ground-breaking study which heralded the
start of the Specialized Approach in foreign language (classroom) anxiety (FLCA)
research. The original definition hints at the complexity of the concept, defining
FLCA as “a distinct complex of self-perceptions, beliefs, feelings and behaviours
related to classroom learning arising from the uniqueness of the language learning
process” (Horwitz, Horwitz and Cope, 1986, p. 128). FLCA was identified as having a
debilitating effect on L2 learning and achievement. The findings have been replicated
in different countries around the globe, with different types of language learners.
More than twenty years later, Horwitz emphasised again that the concept of anxiety is
“multi-faceted” (2010, p. 145). She explained that learners who experience FLCA
“have the trait of feeling state anxiety when participating in language learning and/or
use” (Horwitz, 2017, p. 33).
The idea that negative emotions such as fear, embarrassment, self-doubt, and
boredom hamper progress in L2 development is not new: Krashen (1982) argued that
every learner has an affective filter that determines “the degree to which the acquirer
is "open"” (p. 9). He attributed the idea to Dulay and Burt (1977). When the filter is
“up”, a learner’s understanding and processing of language input would be reduced.
To bring learners’ filters down, teachers were encouraged to spark interest, provide
low-anxiety environments, and bolster learners’ self-esteem (Krashen, 1982, p. 10).
Schumann (1978) had developed a similar concept in his acculturation hypothesis for
SLA. Sufficient contact and social integration with the target language group would
enable a learner to acquire the target language (TL) if “he is psychologically open to
the TL such that input to which he is exposed becomes intake” (p. 29).
Similar ideas were developed outside the field of SLA by Fredrickson (2003),
who reported that negative emotions such as anger lead to the urge to destroy
obstacles in one’s path. However, positive emotions can “broaden people's momentary
thought-action repertoires and build their enduring personal resources, ranging from
physical and intellectual resources to social and psychological resources”
(Fredrickson, 2003, p. 219).
MacIntyre and Gregersen (2012) pointed out that effects of positive emotion go
beyond pleasant feelings: they enhance learners’ ability to notice things in classroom
environment and strengthen their awareness of language input. This, in turn, allows
them to absorb the FL. Positive emotions also help flush out lingering effects of
negative arousal. This is crucial because negative emotions cause a narrowing of
focus and a restriction of the range of potential language input. Positive emotions also
promote students’ resilience and hardiness during difficult times. Crucially, positive
emotion encourages learners to explore and play, two key activities that boost social
Dewaele and MacIntyre (2014) developed a Foreign Language Enjoyment
(FLE) scale consisting of 21 items with Likert scale ratings reflecting positive
emotions towards the learning experience, peers and teacher, which they combined
with 8 items reflecting FLCA. A moderate negative correlation was found to exist
between FLE and FLCA, suggesting that they are partially inter-related but essentially
separate dimensions. Further statistical analysis revealed that those among the 1740
FL learners (from all ages and from all over the world) who were more multilingual,
who had reached intermediate or higher levels in the FL, who felt that were
performing rather better than their peers in the FL class, who were higher up in the
education system (university rather than high school) and who were older,
experienced significantly higher levels of FLE and significantly less FLCA. A
complementary analysis of feedback from 1076 out of the 1746 participants on an
open-ended question related to enjoyable episodes in the FL class showed that
specific positive classroom activities can boost FL learners’ levels of FLE. These
included unusual activities such as debates, making a film or preparing group
presentations. What these activities had in common was that they empowered
students, giving them a choice in shaping the activity so that it matched their concerns
and interests. It confirmed that having a sense of autonomy and having to be creative
enhances performance in the FL. What also emerged from the narratives was that the
classroom environment played a crucial role in the experience of FLE and FLCA.
Participants reported episodes where teachers had been positive, used humour
judiciously, were well-organised, respectful, and praised students for truly good
performance. Sympathetic laughter was appreciated when things went wrong because
it defused a potential negative emotional atmosphere. The narratives showed that
teachers played an important part in their students’ FLE, confirming previous research
(Arnold, 2011). Learners’ feedback also indicated that peers can boost – or destroy -
FLE. Class size was also mentioned, with smaller groups engendering a better
atmosphere, more individual use of the FL and the establishment of closer social
bonds with peers.
A follow-up study by Dewaele and MacIntyre (2016) used a Principal
Components Analysis of the same dataset, and revealed three dimensions explaining a
total of 45% of the variance, and showing the independence of social and private FLE.
FLCA was the first dimension, explaining 26% of the variance, social FLE accounting
for 13% of the variance and Private FLE explaining an additional 6% of variance.
A third study on the same dataset focused on the gender differences at item-
level (Dewaele, MacIntyre, Boudreau & Dewaele, 2016). Statistical analyses
revealed that the 1287 female participants reported having significantly more fun in
the FL class, agreed more strongly that they learned interesting things, and were
prouder of their FL performance than the 449 male peers. They tended to experience
more enjoyment and excitement in a positive FL classroom environment that allowed
them to be creative, and tended to agree more that knowing a FL was “cool”.
However, the female participants worried significantly more than male peers about
mistakes and lacked in confidence in using the FL. No gender difference emerged for
the items reflecting the paralyzing effects of FLCA. The authors argued that the
females’ heightened emotionality might boost the acquisition and use of the FL.
Arnold (1999) attracted the attention of researchers and teachers to the concept
of “affect” in the FL classroom. A growing number of special issues have been
devoted to the topic recently (Arnold, 2011; Avila-López, 2015; Berdal-Masuy &
Pairon, 2015; Puozzo Capron & Piccardo, 2013). Establishing a good emotional
atmosphere in the classroom depends on both learners and teachers and is crucial for
learning to happen. Teachers do play a central role at several levels. They need to
produce comprehensible discourse and create – through verbal and non-verbal means:
“a true learning environment where students believe in the value of learning a
language, where they feel they can face that challenge and where they understand the
benefit they can get from attaining it” (Arnold & Fonseca, 2007, p. 119). Progress in
the FL occurs when good chemistry develops among students, and between students
and their teacher. Good pedagogical practices are crucial to maintain and boost
students' motivation levels and positive emotions (Piccardo, 2013). The FL teacher
needs to use non-threatening techniques in order to create a positive FL learning
experience. This involves supporting and promoting group solidarity and creating an
emotionally safe classroom environment where linguistic experimenting is
encouraged (Arnold, 1999; Baider, Cislaru, & Coffey, 2015; Borg, 2006; Dewaele,
2015; Dörnyei & Csizér, 1998; Dörnyei & Murphy, 2003; Gregersen & MacIntyre,
2014; Williams, Burden, Poulet, & Maun, 2004). Having a positive emotional
atmosphere in a FL classroom is particularly crucial as learners’ self-image is
vulnerable in the FL (Arnold, 2011) and fear of losing face in front of classmates and
teacher can be daunting. Fostering a positive emotional atmosphere can create
linguistic contagion where everybody is “caught” in the FL use (Murphy, 2009).
One crucial aspect in FL classes is that the subject matter needs to be
pertinent, appealing and relevant to FL learners (Arnold, 1999). Developing this
argument, Dewaele (2005, 2011, 2015) has argued that FL classes are too often
emotionally uninteresting or emotion-free which leads to routine, boredom and lack of
engagement. This is not necessarily the teachers’ fault as they are often bound by
strict guidelines about course material and delivery. However, too much rigidity and
overly predictable classroom activities limit the potential for interesting challenges
involving risk-taking, for unpredictability that might cause surprise, and for humour
that boosts enjoyment. Dewaele (2015) pleaded for teachers to have the liberty to do
unexpected, challenging and funny things in their classrooms.
What emerges from the literature review is that although a fair amount of
research has been carried out on language anxiety and FLCA in particular, the study
of positive emotions in the FL classroom merits further investigation. It is crucial to
look simultaneously at both negative and positive emotions because - as Dewaele and
MacIntyre (2016) put it- they are the metaphorical left and right feet of learners on
their way to acquiring the FL. What remains unexplored is to what extent the same
learner-internal and learner-external variables affect both FLCA and FLE within a
specific age-range and a single educational context. This is what the present study
aims to address.
III Research questions
1) What is the relationship between FLE and FLCA?
2) To what extent are FLE and FLCA within one specific educational context linked
to learner-internal variables (age, gender, degree of multilingualism, attitude
towards the FL, level of mastery of the FL, relative standing among peers in the FL
class) and teacher/classroom-specific variables (attitudes towards teacher, frequency
of use of the FL by teacher, time spent reading, writing, listening and speaking in the
FL class and predictability of the FL class)?
3) What are the pedagogical implications of identifying sources of FLE and FLCA?
Languages in UK secondary schools and language attitudes in the UK
The study of a FL is compulsory in UK maintained sector schools at Key
Stage 3 only (students aged 11-14). At Key Stage 4 (students aged 14-16) an FL has to
be offered by the school but it is no longer compulsory (https://www.gov.uk/national-
curriculum/key-stage-3-and-4). FL students in British secondary schools face two national
tests which are high stakes for themselves and for their schools. The results determine
students’ admission into Sixth Form colleges or universities and constitute the basis
for the calculation of national league tables which play a crucial part in the prestige of
the schools. Key Stage 4 students are preparing the national General Certificate of
Secondary Education (GCSE) exams this involves strict exam preparation and a fair
amount of stress for students and teachers. The majority of pupils sitting their GCSE
exams in the UK are 15 years old but at Westminster School three quarters of student
sit their IGCSE1 French a year early at 14. Pupils are under pressure from parents and
school to perform to an expected level and meet targets. FLs are no longer
compulsory at A-level. However, teachers and students are under an equal amount of
pressure as universities typically make conditional offers to secondary school students
who are in their final year, based on students’ personal statement, GCSE results,
predicted A-level results and sometimes university entrance tests and interview
performance. A conditional offer for a language or linguistics degree at Oxford is
typically “AAA” for Cambridge it typically is “A*AA”, meaning a very high score
for three courses. Schools are eager to highlight how many of their former pupils
obtained “A” scores and how many went on to prestigious universities (see footnote
The broader societal context also shapes attitudes towards FL learning and
FLs. Attitudes towards languages are linked to the perception of the group speaking
that language (Gardner, 1985; Mettewie, 2015). In the case of French in the UK,
French is regarded positively. It has traditionally been the most frequently studied FL
in secondary education. As an academic subject it is considered tough and it is highly
recommended to students who wish to go to good universities. The language and its
speakers have an air of sophistication in love, food, culture, and fashion (Dewaele,
2010). France is a popular holiday destination and place of retirement for British
citizens (pre-Brexit). These characteristics of the context are important, because
Gardner pointed out that: “students' attitudes towards the specific language group are
bound to influence how successful they will be in incorporating aspects of that
language” (1985, p. 6).
Participants and Demographics
A total of 189 high school students (49 females, 140 males) participated in the study.
They came from two schools in Greater London: 63 students were from Dame Alice
Owen’s, a semi-selective state school in Potters Bar, and 126 students were from
Westminster School, an independent boarding and day school within the precincts of
Westminster Abbey, which is selective and fee-paying. Both schools are amongst the
top performing schools in the UK2. Dame Alice Owen’s employed 16 full-time and
part-time FL teachers, Westminster School employed 22 full-time and part-time FL
teachers. All students in the study were studying FLs, and 85 students from
Westminster School were also enrolled in courses of Latin and/or Ancient Greek.
Participants’ age ranged from 12 to 18. Eleven were 12 years old, 23 were 13 years
old, 55 were 14 years old, 53 were 15 years old, 18 were 16 years old, 16 were 17
years old and 13 were 18 years old.
A large majority of participants were British (n = 156), often with double
nationalities. Other nationalities included American, Argentinian, Australian, Belgian,
Brazilian, Canadian, Chinese, German, Greek Cypriot, Hungarian, Indian, Iranian,
Irish, Israeli, Italian, Korean, Lebanese, New Zealand, Nigerian, Portuguese, Spanish,
Russian, Singaporean, Swiss and Turkish.
One hundred and sixty-nine students reported to have English as a first
language (L1) with was often combined with other L1s, such as Afrikaans, Arabic,
Bengali, Bulgarian, Cantonese, Dutch, Farsi, French, German, Greek, Gujarati, Hindi,
Hungarian, Italian, Kannada, Korean, Macedonian, Mandarin, Portuguese, Punjabi,
Polish, Russian, Sinhalese, Spanish, Swahili, Tamil, Telugu, Tulu, Turkish and Urdu.
Close to a third of participants (n = 57,) reported growing up with more than one
language from birth.
Most participants were studying French as a FL (n = 144, 68%), while others
were studying Spanish (n = 21), German (n = 15), with smaller numbers studying
Arabic, Dutch, English, Farsi, Hindi, Modern Greek, Italian, Japanese, Mandarin,
Polish, Portuguese, and Russian3.
Participants were also asked about the point they had reached in their FL
journey. Very few described themselves as “beginner” (n = 2), more so as “low
intermediate” (n = 21), “intermediate”, (n = 58), “high intermediate” (n = 92), and
“advanced” (n = 26). The categories of “beginner” and low intermediate” were
Students compared their own FL performance with that of their peers in their
FL class (ranging from “far below average” (n = 1), “below average” (n = 12),
“average” (n = 58), “above average” (n = 92), and “far above average” (n = 26). The
categories of “far below average and “below average” were merged. These values
were positively correlated with self-reported results on their last major FL test (r (187)
= .50, p < .0001). These test scores ranged from 49% to 100%, with a mean of
87.7% (SD = 10). In other words, these were very good FL students.
II The instrument
The questionnaire started with a demographics section from which the above
information was retrieved. Following this, participants were asked to respond to an
item on their attitude towards their first modern FL (as some students learned two FLs
simultaneously), on a 5-point Likert scale. Because very few reported “very
unfavourable” attitudes, this level was merged with the next level, i.e. “unfavourable”
attitudes (n = 18), followed by “neutral” (n = 22), “favourable” (n = 80) and “very
favourable” (n = 69) attitudes. Mean score on the Likert scale was 3.1 (SD = 0.9).
The next question asked whether the student had just one or two FL teachers
for the FL1. Attitudes towards the one - or first- FL teacher were collected using a 5-
point Likert scale (ranging from “very unfavourable” (n = 9), “unfavourable” (n =
10), “neutral” (n = 10), “favourable” (n = 78), to “very favourable” (n = 74) attitudes4.
Mean score on the Likert scale was 4.0 (SD = 1.0).
The following question inquired about frequency of use of the FL in class by
the FL teacher. Answers ranged from “hardly ever” (n = 6) to “not very often” (n =
12), “sometimes” (n = 35), “usually” (n = 77) and “all the time” (n = 59). Mean score
on the Likert scale was 3.9 (SD = 1.0).
The next four questions inquired about the average proportion of time spent on
writing, reading, listening and speaking by the teacher: the options ranged from 0-
10% to 90-100% of the time. Mean scores were highest for writing (36%), followed
by reading (34%), speaking (33%) and listening (32%). In other words, participants
felt that they spent about a third of the time on each of the four skills.
The final question in this section asked how predictable the teacher was during
his/her classes (ranging from “very unpredictable” (n = 3), “unpredictable” (n = 22),
“medium un/ unpredictable” (n = 92), “predictable” (n = 52), to “very predictable” (n
= 13). Because so few participants rated their teacher to be “very unpredictable”, a
single level was created (“very/unpredictable”). Mean score on the 4 point Likert
scale was 2.3 (SD = 0.8).
Students were then invited to complete 10 items, which were extracted from
the Foreign Language Enjoyment questionnaire (Dewaele & MacIntyre, 2014). They
were chosen to capture the reliability of the original scale without sacrificing the
reliability of the measurement. They included items reflecting the three FLE
dimensions: Social FLE, Private FLE and Peer-controlled versus teacher-controlled
positive atmosphere in the FL classroom (Dewaele & MacIntyre, 2016). They were
based on standard 5-point Likert scales with the anchors “absolutely disagree” = 1,
“disagree” = 2, “neither agree nor disagree” = 3, “agree” = 4, “strongly agree” = 5. All
items were positively phrased. A scale analysis revealed high internal consistency
(Cronbach alpha = .88). Mean score for FLE was 3.9 (SD = 0.6).
Another 8 items were extracted from the FLCAS and reflected physical
symptoms of anxiety, nervousness and lack of confidence (Horwitz, Horwitz, & Cope,
1986). They also captured the reliability of the original scale (Dewaele & MacIntyre,
2014). Two FLCA items were phrased to indicate low anxiety and six were phrased to
indicate high anxiety. The low anxiety items were reverse-coded so that high scores
reflect high anxiety for all items on this measure. A scale analysis revealed high
internal consistency (Cronbach alpha = .85). Mean score for FLCA was 2.4 (SD =
The questionnaire was completely anonymous: no names of participants or their
teachers were collected. After the research design and questionnaire obtained approval
from the Ethics Committee of the School of Social Sciences, History and Philosophy
at Birkbeck, University of London, the headmasters of Westminster School and Dame
Alice Owen’s School were contacted to obtain their approval. The first author
presented the research design to the FL teachers at Dame Alice Owen’s, the second
author did the same with his colleagues at Westminster School. Consent was obtained
in two stages: parents were contacted by the school to explain that their children
would be contacted to participate in a survey on affective variables in the foreign
language classroom. They were invited to contact the researchers to obtain extra
information. A couple of parents did so, and none opted out of the survey. Next, the
parents received an email in which they were asked to invite their child to participate
in the study. The student’s individual consent was obtained at the start of the survey.
The questionnaire was posted online using Googledocs.
I The relationship between FLE and FLCA
A Pearson correlation analysis revealed a significant negative correlation between
FLE and FLCA (r (188) = -.194, p < .007). In other words, higher levels of FLE seem
to be linked to lower levels of FLCA but both dimensions share only 3.8% of
variance, a small effect size (cf. Plonsky & Oswald, 2014: 889).
II Learner-internal variables
One-way ANOVAs revealed that age had no effect on FLCA (df (6, 189), F = .6, p =
ns) but did have a significant effect on FLE (df (6, 189), F = 4.0, p < .001, eta2= .118),
Cohen’s d = .73 which represents a medium effect size (Plonsky & Oswald, 2014)5.
Mean scores for participants in each subgroup are presented in figure 1.
Figure 1: The effect of age group on FLE
Two independent t-tests revealed that the 49 female participants scored higher
than the 140 males on FLE: (Females Mean = 4.2, SD = 0.5; Males Mean = 3.8, SD =
0.6). Female participants also scored higher on FLCA: (Females Mean = 2.6, SD =
0.9; Males Mean = 2.3, SD = 0.7). These differences were significant for FLE (df =
187, t = 3.7, p < .0001, Cohen’s d = 0.54, r2= 0.07) and for FLCA (df = 187, t = 2.5, p
< .014 Cohen’s d = 0.37, r2 = 0.03) respectively). According to Plonsky and Oswald
(2014) these are small effect sizes.
One-way ANOVAs showed that the number of languages known to the
participants was unrelated both to FLE and FLCA (df (5, 185), F = 1.6, p = ns and df
(5, 185), F = 1.9, p = .09 respectively).
One-way ANOVAs showed a significant effect of level in the FL on both FLE
and FLCA. More advanced FL learners reported significantly more FLE (df (3, 185),
F = 4.4, p < .005, eta2 = .066, Cohen’s d = .53, a small effect size and significantly
less FLCA (df (3, 185), F = 12.3, p < .0001, eta2= .166, Cohen’s d = .89 - a medium to
large effect size (Plonsky & Oswald, 2014) (see figures 2 and 3). Post-hoc Gabriel
tests6 only showed significant differences in FLE between those who labelled
themselves as low intermediate and intermediate (p < .013). Lower-intermediate
learners reported significantly more FLCA than high intermediate (p < .0001) and
advanced learners (p < .002). High intermediate learners were significantly less
anxious than the low intermediate and intermediate groups (both p < .0001).
Figure 2: The effect of level in the FL on FLE
Figure 3: The effect of level in the FL on FLCA
Relative standing among peers in the FL classroom was also significantly
linked to FLCA (df (3, 185), F = 13.4, p < .0001, eta2= .076, Cohen’s d = .57 - a small
effect size (Plonsky & Oswald, 2014) (see figure 4), but not to FLE (df (3, 185), F =
2.3, p = ns). Post-hoc Gabriel tests showed that participants who felt below average
compared to peers in their class reported more FLCA than those who felt above or far
above average (p < .002 and p < .0001 respectively). Participants who felt average
also suffered more from FLCA than those who felt above or far above average (all p <
Figure 4: Effect of relative standing in the FL classroom on FLCA
The attitude towards the FL was found to have a significant effect on FLE (df
(3, 185), F = 24.8, p < .0001, eta2= .287, Cohen’s d = 1.27, which is a large effect size
(see figure 5), and on FLCA (df (3, 185), F = 4.0, p < .009, eta2= .069, Cohen’s d = .
54 - a small effect size (Plonsky & Oswald, 2014) (see figure 6). Post-hoc Gabriel
tests showed significant differences in FLE between those with (very) unfavourable
attitudes towards the FL and those with favourable (p < .002) and very favourable
attitudes (p < .0001). Those with very favourable attitudes had significantly higher
FLE scores than the three other groups (all p < .0001). Differences in FLCA were
only significant between those with the most negative and the most positive attitudes
towards the FL (p < .019). It was marginally significant between those with
favourable and very favourable attitudes (p = .064).
Figure 5: The effect of attitude towards the FL on FLE
Figure 6: Effect of attitude towards the FL on FLCA
III Teacher-centred variables
One-way ANOVAs showed that attitude towards the teacher had a significant effect
on FLE (df (4, 184), F = 16.7, p < .0001, eta2= .267, Cohen’s d = 1.2 - a large effect
size (Plonsky & Oswald, 2014)- but had little effect on FLCA (df (4, 184), F = 2.3, p
= .06). Unsurprisingly, more positive attitudes corresponded with higher levels of
FLE (see figure 7). Post-hoc Gabriel tests showed significant differences in FLE
between the group of participants with very unfavourable attitudes towards their
teacher and all other groups (all p < .018 or smaller). Similarly, the group of
participants with very favourable attitudes towards their teacher reported significantly
higher FLE than all other groups (all p < .001 or smaller).
Figure 7: The effect of attitude towards the teacher on FLE
The frequency with which a teacher used the FL in class had a significant
positive effect on FLE (df (4, 184), F = 6.2, p < .0001, eta2= .118, Cohen’s d = .73 - a
medium effect size (Plonsky & Oswald, 2014) - but did not affect FLCA (df (4, 184),
F = 0.7, p = ns). In other words, more frequent FL use by the teacher was linked to a
linear increase in levels of FLE among students (see figure 8). Post-hoc Gabriel tests
revealed significant differences in FLE between teachers that used the FL hardly ever
and those who used it usually (p < .036) or all the time (p < .006). Similarly, at the
other end of the scale, students who had teachers that used the FL all the time had
significantly higher FLE scores than students with teachers who used the FL hardly
ever (p < .006), not very often (p < .001) and sometimes (p < .024).
Figure 8: The effect of frequency of teacher’s use of the FL on FLE
The proportion of time that students spent on reading, writing and listening
turned out to be unrelated to levels of FLE and FLCA. However, the amount of time
students spent speaking the FL was positively linked to FLE (df (6, 182), F = 2.5, p
< .023, eta2= 0.076, Cohen’s d = .57 - halfway between a small and a medium effect
size (Plonsky & Oswald, 2014) - but not to FLCA (df (6, 182), F = 0.9, p = ns). Levels
of FLE increased gradually with more time spent on speaking and topped off at 50-
60% of the time dedicated to this activity, after which it dropped off sharply (see
figure 9). However, post-hoc Gabriel tests revealed no significant differences in FLE
between the different groups.
Figure 9: Effect of proportion of time spent on speaking the FL in class on FLE
Teacher predictability was found to have no effect on FLCA (df (3, 185), F = .
2, p = ns) but did have a significant negative effect on FLE (df (3, 185), F = 3.9, p < .
001, eta2 = .06, Cohen’s d = .50 - small effect size (Plonsky & Oswald, 2014) (see
figure 10). Higher levels of predictability were linked to lower levels of FLE. Post-
hoc Gabriel tests revealed only marginal differences in FLE between those who
described their teacher as medium un/predictable and those who found their teacher to
be predictable (p = .058) or very predictable (p = .076).
Figure 10: The effect of teacher predictability on FLE
The first research question addressed the relationship between FLE and FLCA. A
small but significant negative relationship was found, with the dimensions sharing
less than 4% of variance. This confirms the pattern in Dewaele and MacIntyre
(2014), where the amount of shared variance was bigger (13%). The low amount of
shared variance in the present study can be interpreted as further evidence that
enjoyment and anxiety are separate dimensions. In other words, while participants
who score high on FLE will tend to score low on FLCA, it is also possible for
individuals to score high -or low- on both dimensions. Simona, a participant in
Dewaele, MacIntyre, Boudreau and Dewaele (2016), reported such an ambiguous
experience of high anxiety and high enjoyment: “We were supposed to have a 2-
minute speech before our peers and our professor on a topic we chose (…) at first I
was a bit nervous and felt my heart pounding, but it felt great standing there and
expressing my opinion and knowing that all of the other students are listening to you
with attention” (p. 53).
A quick look at the overview of the effects in table 1 shows that learner-
internal variables (e.g., age, gender, FL proficiency levels and attitudes) were more
often linked with FLE and FLCA than the teacher-centred variables (e.g., attitude
towards teachers, teachers’ FL use, predictability), and that the effects of learner-
internal variables were more strongly significant and explained a greater amount of
variance in the two dependent variables (see table 1).
Table 1: Overview of the effects on the independent variables on FLE and FLCA
Variables FLE FLCA
Age *** Ns
Gender *** *
Multilingualism ns Ns
FL level ** ***
Relative standing ns ***
Attitude FL *** **
Attitudes towards teacher *** Ns
Frequency of use of FL by teacher *** Ns
Time spent reading ns Ns
Time spent writing ns Ns
Time spent listening ns Ns
Time spent speaking * ns
Predictability ** ns
*p < .05, ** p < .001, *** p < .0001
The results for age follow the pattern reported in Dewaele and MacIntyre
(2014) with older learners in that study (i.e. those in their thirties, forties, fifties and
sixties) reporting higher levels of FLE than younger learners. The relationship in the
present study was not linear however: the fourteen and fifteen year-olds seemed to
enjoy their FL classes least of all. The explanation may lie with the organisation of
the British FL curriculum. The looming GCSE and A-level exams force teachers to
prepare students for the exams and to “teach to the test”, producing what is referred to
in the literature on assessment and learning as a “negative backwash effect” (see
Ahmad & Rao, 2012). This undoubtedly has a negative effect on FLE. The same
phenomenon could be expected for older, more advanced learners who sit their A-
level exams. However, this effect appears to be overridden by students’ increased
motivation to study the FL, which they have deliberately chosen to pursue at that
stage. This group of students is more likely to enjoy the FL classes. When pupils in
the UK move from eight or nine subjects at age 15 to three or four in their final two
years of schooling, they are generally choosing subjects that they most enjoy and the
skills acquired hitherto are employed in developing greater learner autonomy and an
engagement in more challenging tasks relevant to their chosen path. Other
independent variables undoubtedly affect the overall picture, such as significantly
reduced class sizes, more informal pupil-teacher relationships and greater maturity.
The gender effects uncovered in Dewaele and MacIntyre (2014) and Dewaele,
MacIntyre, Boudreau and Dewaele (2016) were replicated here. The female students
reported both more FLE and more FLCA than their male peers. The findings were
interpreted as an indication that the female learners were more emotionally involved
in the FL learning, experiencing more emotional highs and lows than their male peers.
No significant relationship emerged between the degree of multilingualism
and the dependent variables. This could be linked to fact that the present cohort was
already highly multilingual (an average of 4 languages) so that the knowledge of extra
FLs may have made no further difference. Because this result is difficult to interpret,
it should not be taken to indicate that multilingualism and FLE/FLCA are unrelated
but rather that the study did not produce evidence that they are related.
More experienced FL learners reported both more FLE and less FLCA,
reflecting the finding in Dewaele and MacIntyre (2014). The biggest jump in FLE
and concomitant drop in FLCA was situated between those reporting intermediate and
those reporting high intermediate levels in the FL. One possible explanation is that the
activities that these learners engage in become intrinsically more motivating, as their
newly acquired skills allow them to take on more challenging task including more
The relative standing among peers in the FL class, found to be strongly
positively correlated with the students’ most recent test results, had a significant
negative effect on FLCA, but no effect on FLE. The awareness of not being as good
as the peers is an obvious source of social anxiety. However, feeling stronger than the
peers did not increase participants’ FLE. Dewaele and MacIntyre (2014) found that
higher relative standing was linked to both more FLE and less FLCA.
Unsurprisingly, students who had a more positive attitude towards the FL
reported both significantly more FLE and less FLCA. These students may have
developed a stronger motivation to master the FL and a more burning desire to invest
the time and effort needed to reach that goal.
The aim of the present study was to complement learner-internal variables
with the specific effects of one regional context, namely the teacher and the FL
classroom practices in two Greater London schools. The finding that there is a
positive relationship between positive attitudes towards the teacher and FLE was
expected. A well-loved teacher can boost the enthusiasm of students during classes
(Dörnyei & Csizér, 1998; Gardner, 1985). Enthusiastic students are more likely to
find themselves in a state of flow (Czimmermann & Piniel, 2016; Dewaele, 2015),
which can boost their actual -and self-perceived- performance in the FL. More
surprising was the finding that the attitude towards the teacher was unrelated to
FLCA. In other words, students were equally anxious with much-loved and less-
loved teachers, which suggest that well-loved teachers can create anxiety, but not
necessarily more or less anxiety than other teachers - as indicated by a lack of
significant differences. The present study showed that there are various sources of
anxiety including peer relationships, attitudes, proficiency but previous research has
highlighted the effect of variables, including personality traits, types of evaluations,
curriculum (Dewaele, 2013; Gregersen & MacIntyre, 2014; Horwitz, 2017). Even if
the teachers in the present study did not seem to cause heightened anxiety, some
teachers do, consciously or unconsciously cause anxiety. Those who cause unusually
high levels of anxiety should actually be concerned and adapt their behaviour in an
attempt to reduce its negative effects.
An interesting finding was also that more FL use by the teacher was linked to
more FLE, but not to more FLCA. The debate on the optimal use of the L1 and the
FL in the FL classroom is on-going (Ellis, 2012). There seems to be agreement that
maximising FL use is good wherever possible, but that occasional L1 use can be very
effective (Nation, 2003). Being forced to use the FL can be a source of embarrassment
for beginning and shy learners (Nation, 2003). We do keep in mind that our results
might not apply to the more general FL learner population. Our participants were
very good FL students in selective schools. They were therefore more likely to enjoy
the challenge of having to function in the FL with little L1 use.
Only one type of activity was linked to higher levels of FLE: students who
reported more time speaking the FL also enjoyed their FL classes more, up to sixty
per cent of the time – after which there is a dip. The non-linear relationship indicates
that more is better – up to a point, but that the optimal amount of time spent on
speaking is relatively high. This could be linked to the fact that flow experiences,
which are inherently enjoyable, typically involve speaking in the FL classes with
peers and teacher listening – as the experience of Simona referred to in Dewaele et al.
(2016) illustrated. Listening, reading and writing may be less prone to experiences of
flow. The lack of a statistical effect for these three skills is related to the fact that
there was no clear agreement on what the “optimal” proportion of time would be. The
lack of effects of time spent on the four skills on FLCA shows that how much time
teachers choose to spend on the various skills does not affect how anxious students
feel in the FL classroom.
We are aware that since our research design was non-experimental, we cannot
automatically assume that teachers “cause” variation in FLE because it could be
argued that student behaviour is antecedent. Some students may experience less FLE
because of various reasons. Teachers give grades that reflect the lack of investment,
which further lowers students’ FLE and could boost their FLCA. It is important to
point out that we identified relationships between variables without claiming direct
causality. We also avoided claims about the extent to which FLE or FLCA could
predict un/successful FL learning.
The third research question focused on the pedagogical implications for
teachers of good FL students (as we do not claim that our sample is a representative
sample of the general FL student population). First and foremost, we recommend that
teachers should focus on making their classes enjoyable, because our findings noted a
strong relationship between what teachers actually do in their classrooms and the
extent to which FL students enjoy the FL learning. According to previous literature on
L2 education, there are a number of activities to this end (Dörnyei & Csizér, 1998;
Dörnyei & Kubanyiova, 2014; Dörnyei & Murphy, 2003). For example, it has been
shown that students’ enjoyment can be positively influenced by student-centred
activities where they can have freedom on how to learn the FL in alignment with their
own interests (Dewaele & MacIntyre, 2014). Another crucial technique is to make FL
classroom environments adequately unpredictable, surprising and challenging for
students (Dewaele, 2015).
In contrast, teachers should not be overly concerned about FLCA as they do
not appear to be the main cause of it. Rather, FL students’ anxiety in our sample
seemed to be related to their general FL proficiency and attitude towards FL,
regardless of their immediate FL learning experience with teachers. In fact, FLCA
might be an phenomenon that has unduly monopolized the attention of researchers.
Gardner’s (1985) motivation construct acknowledged the existence of both negative
and positive emotions in the learner – it seems surprising that the negative emotions
have attracted most research attention. Just like the deficit view of learners in the
1960s, when the exclusive focus was on the errors that they produced, researchers on
affective variables have for too long remained obsessed with learners’ negative
emotions. Importantly, our findings suggest that insights from Schumann (1978),
Krashen (1982), and Arnold (1999) about the role of affect were right and that
Positive Psychology can help us further to gain a more balanced view of the emotions
that drive FL (cf. Dewaele, MacIntyre, Boudreau, & Dewaele, 2016; MacIntyre,
Gregersen & Mercer, 2016; MacIntyre & Mercer, 2014).
It is important to mention the fact that a self-selection bias may have skewed
the results towards a more positive portrayal of the FL learning. Although all FL
students in both schools were contacted through their parents, only a fraction filled
out the online questionnaire. It is likely that those who did not feel strongly about
their FL classes were less willing to spend 20 minutes filling out a detailed
questionnaire about their FL classroom experiences. While this may seem like a
drawback, it is in fact also strength. Indeed, Wilson and Dewaele (2010) reported that
the feedback from volunteer FL participants is of much better quality than from
participants who were forced to fill out a questionnaire. Our findings may thus not be
generalised to the whole FL population, but it gives us a very good panorama of those
who probably best described as “good language learners”.
Our British secondary school learners reported significantly higher levels of FLE than
FLCA, with a weak negative relationship between both. It confirms the observation
that FLE does not represent the positive pole of some generic emotion dimension with
anxiety at the negative pole. FLE and FLCA are independent emotion dimensions
with a very small amount of overlap. This means that students could score high, or
low, on both dimensions. The pedagogical implication is that teachers’ attempts to
reduce FLCA will not automatically boost student’ FLE.
Positive attitudes towards the FL, the FL teacher, a lot of FL use by the teacher
in class, a strong proportion of time spent by students on speaking, a higher relative
standing among peers in the FL class and a relative advanced stage of development of
the FL all contributed to higher levels of FLE.
Fewer variables were linked to lower levels of FLCA, namely positive
attitudes towards the FL, higher relative standing among peers in the FL and a relative
advanced stage of development of the FL. The striking difference between the
dimensions of FLE and FLCA is that the latter dimension seems much less related to
teacher and teacher practices than FLE.
We conclude that effective teachers fuel learners' enthusiasm and enjoyment
and do not spend too much time worrying about their FLCA. This includes the
creation of friendly low-anxiety environments without fixating on single negative
emotions. Metaphorically, we suggest that teachers should seek to light the students’
fire by being engaging, by creating interest in the FL and by using it a lot in class
rather than worry too much about students feeling cold. Once the right emotional
temperature is reached, students will forget about the cold and will jump into action,
reaching their own optimal temperature. The advice is not new but the evidence to
support the argument certainly is.
We would like to thank Dr Davison, the Headteacher, the Head of Languages, Ms
Davies and the teachers of Dame Alice Owen’s as well as the Head Master of
Westminster School, Mr Derham, the Head of Languages and the language teachers
for allowing us to contact their FL students. Thanks also to the students’ parents and
the students themselves for their collaboration in this project.
Ahmad, S., & Rao, C. (2012). A review of the pedagogical implications of
examination washback. Research on Humanities and Social Sciences, 2, 11-20.
Arnold, J. (1999). (Ed.) Affect in Language Learning. Cambridge: Cambridge
Arnold, J. (2011). Attention to affect in language learning. Anglistik. International
Journal of English Studies, 22, 11-22.
Arnold, J., & Fonseca, C. (2007). Affect in teacher talk. In B. Tomlinson (Ed.),
Language Acquisition and Development (pp. 107-121). London: Continuum.
Avila-López, J. (2015). Didáctica de la emoción: De la investigación al aula de ELE.
MARCO ELE: Revista de Didáctica Español Lengua Extranjera, 21, 1-71.
Baider, F., Cislaru, G. & Coffey, S. (2015). Apprentissage, enseignement et affects. Le
langage et l’homme, 50 (L1), 3-6.
Berdal-Masuy, F. & Pairon, J. (2015). Affects et acquisition des langues. Le langage
et l’homme, 50 (L2), 3-8.
Borg, S. (2006). The distinctive characteristics of foreign language teachers.
Language Teaching Research, 10, 3-32.
Chastain, K. (1975). Affective and ability factors in second language acquisition.
Language Learning, 25, 153-161.
Czimmermann, E., & Katalin P. (2016). Advanced language learners’ experiences of
flow in the Hungarian EFL classroom. In P. D. MacIntyre, T. Gregersen, & S.
Mercer (Eds.), Positive Psychology in SLA (pp. 193-214). Bristol: Multilingual
Dewaele, J.-M. (2005). Investigating the psychological and the emotional dimensions
in instructed language learning: Obstacles and possibilities. Modern Language
Journal, 89, 367-380.
Dewaele, J.-M. (2009). Individual differences in Second Language Acquisition. In W.
C. Ritchie & T. K. Bhatia (Eds.), The New Handbook of Second Language
Acquisition. Bingley (UK): Emerald, pp. 623-646.
Dewaele, J.-M. (2010). The perception of French by native speakers and advanced L2,
L3 and L4 learners. In V. Regan & N. Chasaide (Eds.), Language Practices and
Identity Construction in French (pp. 133-156). Bern: Peter Lang.
Dewaele, J.-M. (2011). Reflections on the emotional and psychological aspects of
foreign language learning and use. Anglistik. International Journal of English
Studies, 22, 23-42.
Dewaele, J.-M. (2013). The link between Foreign Language Classroom Anxiety and
Psychoticism, Extraversion, and Neuroticism among adult bi- and multilinguals.
The Modern Language Journal, 97 (3), 670-684.
Dewaele, J.-M. (2015). On emotions in foreign language learning and use. The
Language Teacher, 39, 13-15.
Dewaele, J.-M., & MacIntyre, P. D. (2014). The two faces of Janus? Anxiety and
Enjoyment in the Foreign Language Classroom. Studies in Second Language
Learning and Teaching, 4, 237-274.
Dewaele, J.-M., & MacIntyre, P. D. (2016). Foreign Language Enjoyment and
Foreign Language Classroom Anxiety. The right and left feet of FL learning? In
P. D. MacIntyre, T. Gregersen, & S. Mercer (Eds.), Positive Psychology in SLA
(pp. 215-236). Bristol: Multilingual Matters.
Dewaele, J.-M., MacIntyre, P. D., Boudreau, C., & Dewaele, L. (2016). Do girls have
all the fun? Anxiety and Enjoyment in the Foreign Language Classroom. Theory
and Practice of Second Language Acquisition, 2 (1), 41–63.
Dörnyei, Z., & Csizér, K. (1998). Ten commandments for motivating language
learners: Results of an empirical study. Language Teaching Research, 2, 203-229.
Dörnyei, Z., & Kubanyiova, M. (2014). Motivating learners, motivating teachers:
Building vision in the language classroom. Cambridge: Cambridge University
Dörnyei, Z., & Murphy, T. (2003). Group dynamics in the language classroom.
Cambridge: Cambridge University Press.
Dörnyei, Z., & Ryan, S. (2015). The psychology of the second language learner
revisited. New York: Routledge.
Dulay, H., & Burt, M. (1977). Remarks on creativity in language acquisition. In M.
Burt, H. Dulay, & M. Finnochiaro (Eds.) Viewpoints on English as a Second
Language (pp. 95-126). New York: Regents.
Ellis, R. (2012). Language teaching research and language pedagogy. Chichester,
Field, A. P. (2013). Discovering statistics using IBM SPSS Statistics: And sex and
drugs and rock 'n' roll. London: Sage (4th ed.).
Fredrickson, B. L. (2003). The value of positive emotions. American Scientist, 91,
Gardner, R. C. (1985). Social psychology and second language learning: The role of
attitudes and motivation. London: Edward Arnold.
Gregersen, T., & MacIntyre, P. D. (2014). Capitalizing on Individual Differences:
From Premise to Practice. Bristol: Multilingual Matters.
Horwitz, E. K. (2010). Foreign and second language anxiety. Language Teaching, 43,
Horwitz, E. K. (2017). On the misreading of Horwitz, Horwitz, and Cope (1986) and
the need to balance anxiety research and the experiences of anxious language
learners. In C. Gkonou, M. Daubney & J.-M. Dewaele (Eds.), New Insights into
Language Anxiety: Theory, Research and Educational Implications (pp. 31-50).
Bristol: Multilingual Matters.
Horwitz, E., Horwitz, M., & Cope, J. (1986). Foreign language classroom anxiety.
Modern Language Journal, 70, 125-132.
Kleinmann, H. (1977). Avoidance behavior in adult second language acquisition.
Language Learning, 27, 93-107.
Krashen, S. (1982). Principles and practice in second language acquisition. Oxford:
MacIntyre, P. D. (2017). An overview of language anxiety research and trends in its
development. In C. Gkonou, M. Daubney, & J.-M. Dewaele (Eds.), New insights
into language anxiety: Theory, research and educational implications (pp. 11-30).
Bristol: Multilingual Matters.
MacIntyre P., Gregersen T., & Mercer S. (Eds.) (2016). Positive Psychology in SLA.
Bristol: Multilingual Matters.
MacIntyre, P. D., & Gregersen, T. (2012). Emotions that facilitate language learning:
The positive-broadening power of the imagination. Studies in Second Language
Learning and Teaching, 2, 193-213.
MacIntyre, P. D., & Mercer, S. (2014). Introducing positive psychology to SLA.
Studies in Second Language Learning and Teaching, 4, 153-172.
Mettewie, L. (2015). Apprendre la langue de « l’Autre » en Belgique: la dimension
affective comme frein à l’apprentissage. Le Langage et l’Homme, L2, 23-42.
Murphy, T. (2009). Some crucial elements of learning ecologies of linguistic
contagion. In G. Goncalves et al. (Eds.) New Challenges in Language and
Literature (pp. 129-147). Bello Horizonte: Faculdade de Letras da UFMG.
Nation, P. (2003). The role of the first language in foreign language learning. Asian
EFL Journal, 5, 1-8.
Piccardo, E. (2013). Évolution épistémologique de la didactique des langues: la face
cachée des émotions. Lidil, 48, 17-36.
Plonsky, L., & Oswald, F. L. (2014). How Big Is “Big”? Interpreting effect sizes in L2
research. Language Learning, 64, 878–912.
Puozzo Capron, I. & Piccardo, E. (2013). “Au commencement était l’émotion”:
Introduction. Lidil, 48, 5-16.
Schumann, J. (1978). The acculturation model for second-language acquisition. In R.
Gingras (Ed.), Second-Language Acquisition and Foreign Language Teaching
(pp. 27-50). Arlington, Virginia: Center for Applied Linguistics.
Scovel, T. (1978). The effect of affect on foreign language learning: A review of the
anxiety research. Language Learning, 28, 129-142.
Williams, M., Burden, R.L., Poulet, G., & Maun, I. (2004). Learners' perceptions of
their successes and failures in foreign language learning. Language Learning
Journal, 30, 19-29.
Wilson, R., & Dewaele, J.-M. (2010). The use of web questionnaires in second
language acquisition and bilingualism research. Second Language Research, 26,
1 The International General Certificate of Secondary Education (IGCSE) is an English language
curriculum offered to students to prepare them for International Baccalaureate, A Level and BTEC
Level 3 (which is recommended for higher tier students).
2 Dame Alice Owen’s School reported that 81% of all grades were awarded A* - B at A-level in
2015 (with 205 students participating in the exams).
Westminster School reported that 97% of all grades were awarded A* - B at A-level in 2015 (with
583 students participating in the exams).
3 The rank order corresponds to national figures for the 23,031 A-level entries in the UK in 2015,
with 45% of students choosing French, followed by Spanish (38%) and German (17%)
4 Because of the anonymity it is impossible to know how many different teachers participants
commented on. Considering that there were sufficient numbers of participants from all year groups
in both schools, we assume that our participants provided us with objective reports of actual teacher
behaviour of close to 38 teachers.
5 The authors suggest the following: “we (…) urge L2 researchers to adopt the new field-specific
benchmarks of small (d = .40), medium (d = .70), and large (d = 1.00) in order to interpret the
practical significance of L2 research effects more precisely” (2014, p. 889).
6 Field (2013) recommends the Gabriel post-hoc tests when sample sizes are different because they
have greater power.