| This Issue | Articles | Instructional Ideas | Open Forum | Reviews | Authors |
| Dentith | Lee | Morgan | Ross |

[ Art Reviews | Book Reviews | Multimedia Reviews ]

Gender Effect on Error Treatment in University ESL Classrooms 

 Jiyoon Lee
University of Pennsylvania
U. S. A.

Introduction
Literature Review
Methods
Results
Discussion
References

Introduction

Error treatment is an important issue in language learning and teaching.  Language teachers have ample experiences in treating their students’ errors.  When I taught high school students in Korea, I had many opportunities to correct errors that students made while learning English.  Since I was aware that public correction could cause embarrassment to adolescent students, I was careful in my ways of treating errors.  I especially recall that girls were more embarrassed when being corrected in front of others than boys.  Therefore I adopted different ways of dealing with their oral mistakes depending on students’ gender.  I have been wondering if there is any relationship between students’ gender and teacher-initiated error treatments beyond my personal experience. My research focuses on two aspects of error treatment: (1) the patterns of teachers' error treatment depending on the students' gender and (2) students’ preference and perception of teacher-initiated error treatment.  Provided that treating error is one of the most significant and delicate matters in language teaching, the results of this study could assist teachers with effectively treating students' errors while causing minimal anxiety.  [paragraph 1]

   Literature Review

In early studies of language and gender, some researchers assumed that men and women have a tendency to use different styles of language.  Lakoff’s (1973) study of women’s preference for certain patterns of language use exemplifies the earlier studies that tend to oversimplify the relationship between language and gender.  According to Lakoff, women often express themselves hesitantly, tentatively, weakly, trivially, and politely.  These characteristics of women’s language are manifested in their selection of lexicon, syntax, phonology, and prosody. As studies went on, researchers began to seek empirical evidence of gender differences and to examine the contexts and speech community in which men and women talk.  Trying to understand the importance of contexts of interaction between men and women, some researchers like Leap (1999) suspected that people become "gendered" as they live in the society.  He identified media and interaction with peer groups as sources of gender identity construction in his article, “Language, Socialization, and Silence in Gay Adolescence.”  School is considered as one of the most crucial places where students' identities, including gender identity, develop as they spend a large number of hours daily in a school environment.  [paragraph 2]

Understanding the importance of school settings in identity construction, many studies were conducted with the focus of gendered phenomena in classrooms, revealing the effect of the gender of teachers and students on their mutual interactions.  Good, Sikes, and Brophy (cited in Lee, Marks, and Byrd, 1994) found in the study of sixteen junior high teachers that both male and female teachers gave more attention to boys and that the intensity of academic interaction of teachers was qualitatively different with boys than with girls.  Since higher-level thinking questioning was identified as a critical element in effective teaching and learning, much research in qualitative gender biases has been focused in that direction. [paragraph 3]  

Several studies support the argument that the gender of students affects teachers' interaction with them.  Sadker and Sadker's (1990) study of gender equity in the classroom revealed that teachers subconsciously gave qualitatively different evaluation and feedback to male students than to female students.  Not only male students get more attention but also more criticism than female students (Dweck and Gilliard, 1975; Brothy, 1985).  Dweck and Gilliard noted that teachers tended to criticize boys’ poor academic performance for not trying hard enough whereas this form of criticism was rarely directed toward girls.  Brothy argued that under the same instructional approaches teachers criticized boys more frequently and gave girls more praise. Female students do not always fare better. Barnes (1990) discovered that male and female teachers' written comments on their students' essays varied depending on the students' gender.  Male teachers were more intolerant of emotional writing and more critical when the author was female; female teachers tended to be more fastidious about mechanics and more concerned about the form of an essay.  [paragraph 4]

Fagot’s (1981) study emphasized teachers’ experience more than the gender of teachers and students in preschool settings: experienced teachers initiated more interactions with students--both boys and girls--when they were engaged in female-preferred activities such as playing in the kitchen, playing with dolls, and engaging in art activities  This was found to be true for both male and female experienced teachers.  In contrast, less experienced teachers were observed to interact more with boys when they were engaged in similar activities.  They asked more questions and provided more information to girls but joined boys' on-going play activities more often.  [paragraph 5]

Methods

As stated above, much research has been conducted in mainstream classroom settings.  When considering the growing number of immigrant and international students learning English in the United States, however, the necessity of studying the interaction patterns between ESL instructors and these students surfaces.  This study addresses this less explored area with two questions related to the gender effect on teacher-student interaction: (1) Does student gender affect the frequency and types of teacher-initiated oral error treatment and (2) Does student gender affect their reception of error treatment?   Based on the literature, it is hypothesized that oral errors students make in their ESL classes will be treated differently in frequency (quantitatively) and in type (qualitatively) by teachers depending on student gender.  Two sub-hypotheses were developed: (1) female students will receive less error treatments than male students; and (2) the errors female students make will be treated in a less explicit manners than male students. [paragraph 6]

This observational study was conducted in five university ESL (English as a Second Language) classrooms.  These observation sites were selected according to the criterion that the goal of the classroom is to improve students’ speaking ability.  The selected classrooms were developed to help students improve accuracy as well as fluency of their English speaking proficiency and covered all levels: the beginner (1 class),  intermediate (2), high-intermediate (1), and advanced level (1).  All classes were taught by female native-English-speaking instructors whose ages ranged from the 20's to the 40's.  Fifty-five students from these classes, whose average age is 25, participated in the study during the Spring semester of 2002.  Thirty of them were male and twenty-five female; more than eighty five percent of the subjects came from Asian countries such as Korea, Japan, and Taiwan. [paragraph 7]  

Three methods of data collection were adopted: classroom observation, survey of students, and interviews of instructors.  During the classroom observation data on the frequency and type of error treatments initiated by instructors was collected.  This proved to be the most crucial method of data collection because this allowed the data to be contexualized in terms of student gender and circumstances in which errors were corrected.  The survey was also necessary to gather data on the effect of students' gender on their preference for certain types of error treatments.  The interviews of instructors after some lessons yielded valuable results in helping to understand whether there is any discrepancy between what they intended to do and what they actually did in classrooms.  [paragraph 8]

The collected data was analyzed in two ways.  First, the quantity (frequency) of teacher-initiated error treatments was measured by counting the number of error treatments that teachers gave during the observations.  Second, the quality (type) of teacher-initiated error treatments was determined by counting the number of error treatments that teachers gave in each predetermined category: explicit treatment, recast, clarification request, metalinguistic feedback, or repetition. According to Lyster and Ranta (1997), an explicit treatment refers to the activity that the teacher points out a student's incorrect utterance and provides a correct form. Recast means that without a direct acknowledgement of student's errors the teacher implicitly reformulates the student's error or provides the correction. Clarification request is expressed by phrases like "Excuse me?" or "I don't understand."  The request implies that repetition or reformulation is required because the message has not been understood or the student's utterance contained some mistakes. In metalinguistic feedback the teacher poses questions or provides comments or information related to the formation of the student's utterance without providing the correct form. Finally, repetition means that the teacher repeats the student's error and adjusts intonation to draw student's attention to it. At the beginning of each observation, a diagram of students’ seating position was drawn and the gender of each student was marked.  Occurrences of error treatments were then recorded for each student, which later helped to determine the frequency and type of error treatments by students'  gender.  To secure the reliability of collected data, the focus of the observation was not disclosed to either instructors or students during the observation.  [paragraph 9]

Results

The total of seventy-two teacher-initiated oral error treatments was gathered during five class periods.  Forty-seven of the treatments (65%) were directed to male students  whereas 25 of these were directed to female students.  Provided more female students volunteered to answer questions, error treatments were given to male students twice as much as to female students (.72 treatments to males vs. .35 treatments to females) per voluntary participation (See Table 1).  

Table 1. Teacher-initiated Error Treatments by Students' Gender

[paragraph 10]  

Most of the error treatments were given in the form of explicit treatments or clarification request, followed by  metalinguistic explanation, recast, and repetition (see Table 2).  The following examples illustrate the explicit treatment:

Example 1:

Student: She come soon.

Teacher: Instead of “come”, you should use “is coming.” She is coming.

Example 2:

S: Yes, I don’t know.

T: Hmm, you said, “Yes, I don’t know." [explanation] So you should say either “Yes, I know.” or “No, I don’t know.” Otherwise [explanations].   

Male students received overall more explicit treatments (45%) whereas female students received more clarification requests (40%) which is a weaker version of explicit treatment.  In some classes almost ninety percent of all error treatments was aimed at male students either in the form of explicit treatment or clarification request.   

Table 2. Types of Error Treatments by Gender

[paragraph 11]  

While instructors' error treatments depended on student gender, the gender of students did not affect their perception of teacher-initiated error treatments.   In response to a likert-type questionnaire containing eight "I like..." statements both male and female students expressed positively their liking of receiving all types of error treatments by instructor (see Table 3). Students preferred explicit error treatments and grammatical explanation to recast and repetition. Interestingly female participants agreed more with explicit oral error treatments (100% vs. 93%) and metalinguistic error treatments such as grammatical explanation (92% vs. 80%) than male counterparts. 

Table 3. Students' Preference of Teacher-initiated Error Treatments

Female students did not like the idea of having their errors corrected after class or via e-mail, which may mean they prefer immediate feedback from instructors.  Participants in this research, regardless of their gender, prefer in-class error treatments to after-class feedback.  However, in terms of using e-mail, gender difference was visible: male students seemed to have a more ambivalent opinion of this format indicated by even distribution of answers whereas the majority of female students (54%) disagreed with this form of delayed error treatments.  [paragraph 12]

Instructor interviews revealed that those who were interviewed mentioned that they did not treat students’ errors differently depending on students’ gender but considered students’ age and their native culture.  [paragraph 13]

 Discussion

The results of this study support the conclusion of Sadker & Sadker's (1990) study:  male students received more error treatments in explicit ways although instructors did not intend to differentiate them or were aware of their gender-based pattern.   Interestingly, however, students’ gender difference did not affect their preference of error treatments:  both groups liked the idea of having their errors treated explicitly during class.  [paragraph 14]

The study showed that the gender difference of students is related to teachers' ways of treating students' errors.  Yet the gender-differential treatment needs to be explored in depth, rather than simply being relegated to gender bias of the instructors.  The teacher's differential treatment of male and female errors could be accounted for by several possible explanations.  It is possible that female students made fewer errors or less critical errors that required teachers' attention or that they might have resorted to other sources of error treatments such as self or peers, which made teacher intervention unnecessary.  It is also possible that the gender difference in error treatment is only in quality, not in quantity.   Furthermore, classroom activities and instructors' personal preference should be taken into consideration.  Certain activities in classrooms--such as new vocabulary check--involved more error treatments and some instructors preferred to utilize certain types of error treatments on the day of observation.  [paragraph 15]

Despite the limitedness of data collection and interpretation in this study, the findings support the existing literature on the gender-based student-teacher interaction.   As many researchers concluded in previous studies, it is important to consider that gender related patterns are deeply intertwined with other dynamics of language behaviors.  Along with gender differences, people's language learning and usage are affected by ethnicity, class, regional origin, age, professional training, and individual personality.  In terms of error treatments, we as language teachers should also remember that individual learners may have different preferences for certain error treatments styles.  Thus it would be wise for language teachers to learn about their students.  [paragraph 16]

References

Barnes, L. (1990). Gender bias in teachers’ written comments.  In S. Gabriel & I. Smithson. (Ed.), Gender in the classroom (pp140-159). Chicago: University of Illinois. 

Brophy, J. (1985). Interactions of male and female students with male and female teachers. In L. Wilkinson, L. & C. B. Marrett (Ed.), Gender influences in classroom interaction (pp. 115-142). New York: Academic Press.

Dweck, C. & Gilliard, D. (1975). Expectancy statements as determinants of reactions to failure: Sex difference in persistence and expectancy change. Journal of Personality and Social Psychology, 32, 1077-1084.

Fagot, B. (1981). Male and female teachers: Do they treat boys and girls differently? Sex Roles; A Journal of Research, 7 (3), 263-271.

Lakoff, R. (1973). Language and woman’s place. Language in Society, 2, 45-79.

Leap, W. (1999). Language, socialization, and silence in gay adolescence. In M. Bucholtz, A. C. Liang, and L. A. Sutton.  (Eds.).  Reinventing identities: The gendered self in discourse (pp. 259-272).   New York: Oxford University.

Lee, V. E., Marks, H. M., & Byrd, T, J. (1994). Sexism in single-sex and coeducational independent secondary school classrooms. Sociology of Education, 67 (2), 92-120.

Lyster, R. & Ranta, L. (1997). Corrective feedback and learner uptake: Negotiation of form in communicative classrooms.  Studies in Second Language Acquisition, 19, 37-66.

Sadker, M. & Sadker, D. (1990). Confronting sexism in the college classroom. In S. Gabriel & I. Smithson. (Eds.), Gender in the classroom (pp.176-187). Chicago: University of Illinois.

Jiyoon Lee ,. is pursuing her M. Ed. in TESOL at the University of Pennsylvania. She has taught English in ESL and EFL settings.  Her research interests include input characteristics which promote and facilitate L2 learning. (Contact the author at jiyoon@dolphin.upenn.edu) 

Lee, Jiyoon. (2002). Gender effect on error treatment in university ESL classrooms . Electronic Magazine of Multicultural Education [online], 4 (2), 16 paragraphs <Available: http://www.eastern.edu/publications/emme/2002fall/lee.html> [your access year, month date]