The influence of gender and self-efficacy on the English writing ability of central Indonesian students

Gender and Self-Efficacy are two things that can influence the quality of student learning in the classroom. This applied linguistics study investigates the impact of gender and self-efficacy on Indonesian students' English writing proficiency. This study aims to explore potential differences in writing abilities between male and female students and the influence of their self-efficacy on their writing performance. It used a mixed methods research design that combined quantitative measures and qualitative interpretation. It selected some Indonesian students from several educational institutions as the respondents. The respondents’ writing abilities were assessed through a standardized writing test. In addition, a questionnaire related to self-efficacy was conducted to measure students' self-efficacy and its correlation to their writing abilities. The quantitative analyses revealed important differences in writing performance based on gender and varying levels of self-efficacy. In addition, the qualitative findings provide insight into the factors that influence students' self-efficacy and their perceptions of writing as a skill. The implications of this study draw attention to the importance of addressing individual differences and self-perceptions in language learning contexts. Understanding the influence of gender and self-efficacy on English writing skills can direct educators in adapting learning approaches to effectively improve students' writing skills. The study concludes that English writing abilities among the students reveals significant gender-based influences on the development of skills: male students' scores were lower than female students', highlighting gender-based disparities. Self-efficacy also played a significant role in enhancing students' English writing abilities. It also shows that there is a significant interaction effect between gender and self-efficacy. It affects the trajectory of English writing skills. There are many factors influencing the development of English writing abilities that calls for further exploration in academic discourse.

collaborate and communicate are among the targets of the curriculum.Therefore, the main objective of the curriculum is to give graduates a well-rounded education and prepare them for life after graduation.
In learning English, the new curriculum also implies the use of appropriate learning models.This is because there are four different competencies: writing, speaking, listening, and reading.This situation has a major impact on students' learning strategies (see Anggarista & Wahyudin, 2022) because the aim of learning English in the context of higher education in Indonesia is to develop language skills, both orally and in writing.The language skills in question are the ability to listen, speak, read and write.
Most Indonesian students consider writing, especially in English, to be very difficult and the learning process is boring (Kemalsyah et al., 2022;Muamaroh et al., 2020).The difficulties experienced by students can be seen when they must transfer ideas from Indonesian to English.The second difficulty is the student's inability to determine the meaning of words or phrases in a piece of writing.Apart from that, the learning process is still traditional, namely placing more emphasis on the results of students' writing than on the process that should be carried out (Khair & Misnawati, 2022).Students are immediately asked to write without learning how to write.Lecturers usually give several types of topics and ask students to choose one, then they immediately write (cf.Handayani & Aminatun, 2020).
Traditional methods tend to ignore the importance of the process of learning to write, which should include developing ideas, planning, and editing (Bulqiyah et al., 2021;Yang et al., 2021).This approach also pays little attention to the importance of linguistic understanding and choosing the right words, which are often the main obstacles for students in English writing (cf.Hassan et al., 2020).Therefore, students not only feel overwhelmed but also lack confidence in their writing skills.To overcome this problem, a more comprehensive and process-oriented learning approach is needed, which does not only focus on the result but also on the process of developing the writing skills.
Several studies have identified problems that foreign language learners often face when learning to write in English.Some of these include a lack of facility to express ideas (Budjalemba & Listyani, 2020), the long time it takes to write (Sun & Wang 2020), lack of vocabulary (Ismayanti & Kholiq, 2020), and the influence of the first language (Hassan et al., 2020).Grammatical aspects are also a serious concern in learning to write, especially in matters that greatly influence the meaning and quality of writing academically (Crossley, 2020).
On the other hand, a teacher must be able to provide appropriate correction to writing errors, considering several things such as correction must be selective and carried out based on student development consistently and effectively (Ghasemi, Noroozi, & Salehan, 2021).An example of selective correction is that a teacher must be able to choose correctly which errors need to be corrected immediately and which ones do not, so that the corrections made do not hinder the delivery of the material and do not interfere with the learning process (Jinowat & Wiboolyasarin, 2022).This approach is important because not all errors require immediate intervention.Excessive attention to every small error can reduce a student's self-confidence and slow down the learning process.Focusing on the most critical errors can help teachers to improve fundamental aspects of students' writing, while still allowing the learning process to run smoothly and effectively (see Yu & Liu, 2021).
Correction has the same function as feedback.Success or failure in providing feedback can be influenced by several factors, namely the classroom context, type of error, student ability level, and type of writing (Yunus, 2020).Apart from that, identifying student learning styles or models by teachers is very important because it can help students more easily follow the learning process.This can also contribute to the development of students' abilities (Lwande et al., 2021).Therefore, apart from the teacher's ability to teach, an appropriate learning model is also needed to develop students' English writing skills.Understanding students' learning styles allows teachers to adjust teaching methods to be more effective, for example by using visual, auditory or kinesthetic approaches according to students' learning preferences (Alhourani, 2021).Thus, feedback and correction not only serve to correct errors, but also to improve the overall learning process, provide a motivational boost, and build students' confidence in writing English.
Learning to write with appropriate meaning is considered the most difficult language skill compared to other language skills, both in the first or mother tongue, or as a foreign language (Kahveci & Şentürk, 2021).Writing is regarded as the most challenging talent for learners, particularly in cases when English is a foreign language that is challenging to learn.The challenge is not just in structuring and refining concepts, but also in converting them into readable text (Rets et al., 2022).
Students face complex problems in learning writing, including difficulty on understanding sentences due to grammatical and mechanical errors (Muamaroh et al., 2020), low motivation (Ali & Zayid, 2022), difficulty on gathering ideas (Pham 2021), and individual learning (Arifani et al., 2020).Learning is a continuous process influenced by experience, interaction with the world, and technological developments (Chuang, 2021).
Language learning styles and strategies are important variables that influence second language learners' abilities.Academic writing, particularly, requires serious effort and practice, including prewriting, writing, revising, and editing stages.Understanding the function of "to be" in various sentence forms and recognizing individual words are also crucial in English writing activities.
Many studies that have examined writing courses in higher education context (Budianto et al., 2020;Imran & Almusharraf, 2023).Some of them have discussed the relationship between selfefficacy and English writing skills (Dai et al., 2023;Mitchell et al., 2023).In addition, many studies have focused on the context of students who use English as their second language or in environments that support intensive English use (Pilotti et al., 2023;Sun & Wang, 2020;Teng & Zhang, 2020).However, studies on how self-efficacy or gender affects writing skills in the context of students learning English as a foreign language have always been interesting (cf.Grenner et al., 2021).The importance may be especially in countries with limited access to English-speaking environments (see Sun et al., 2021).Such study is important because students in non-native environments have additional challenges, such as limited language practice (Grenner et al., 2021) and exposure to diverse language models (Shen & Bai, 2024), which can affect their perceptions of self-efficacy and their learning outcomes.
Indeed, most studies on writing-learning tend to emphasize the cognitive and technical aspects of developing writing skills, such as grammar and text structure, without paying close attention to psychological and affective factors such as selfefficacy (Yu et al., 2023).In fact, self-efficacy plays an important role in determining how much effort and persistence students put into complex writing tasks.This provides additional opportunities for investigating the potential mediating or moderating effects of self-efficacy on the link between other emotional components and writing learning outcomes (Yu et al., 2023).It also raises the question of how to create instructional interventions that will boost students' writing abilities and selfefficacy.It becomes pertinent when considering English language instruction in underdeveloped nations, where there are frequently insufficient resources and opportunities for language skill development (Akram et al., 2020).
Based on these considerations, this study focused on a writing course model at a university in central Indonesia.The course aims to develop language skills with more emphasis on writing abilities and includes an understanding of text structure, correct use of English grammar, as well as the development of ideas and arguments in academic writing.Thus, this course not only aims to improve technical skills in writing, but also to equip students with the critical and analytical thinking skills needed in academic writing.

METHOD
This study used an experimental method with a treatment by level 2 x 2 design.The variables consist of one dependent variable, namely the result of learning to write English and two independent variables, namely the learning model as the treatment variable and self-efficacy as the moderator variable.The learning model variables consisted of self-regulated learning models and direct learning.Meanwhile, the self-efficacy variable consists of high and low self-efficacy.Figure 1 explains the scheme of this study that seeks to explore the interaction between learning models and selfefficacy on student learning outcomes in writing English, as well as determining the possibility of more effective learning models based on student self-efficacy levels.

Figure 1 Treatment Design by Level 2x2
Information: A1B1 = Groups of students who were taught to write English using the Self-Regulated Learning model for students who have high self-efficacy abilities A2B1 = Group of students who were given English writing lessons using a direct learning model for students who have high self-efficacy.A1B2 = Group of students who were taught to write English using a self-regulated learning model that has low self-efficacy abilities.A2B2 = A group of students who were given English writing lessons using a direct learning model that had low self-efficacy abilities This study was conducted at a university in central Indonesia, based on the schedule of writing courses in an even semester.The study lasted for one semester with a total of 8 meetings, each meeting with a duration of 2 x 50 minutes.The duration of each meeting follows the curriculum provisions that apply in universities in Indonesia, where one hour is equivalent to 50 minutes.In the curriculum, this course has a burden of 2 credits, so that each meeting lasts for 2 x 50 minutes.
The target population in this study was students majoring in English education.The population included students who took the second level writing course in the odd semester of the 2017/2018 academic year.It consists of two classes: class A with 60 students and class B with 62 students.
The sampling is done using a simple random sampling technique.This random sampling technique aims to provide the equal opportunity for each individual in the population to be selected as a sample (Nanjundeswaraswamy & Divakar, 2021).It was expected that it would increase the external validity of the study and minimizes selection bias.
Two classes were randomly selected for this study, using draw letters A and B. The draw show class B as the experimental class and class A as the control class.Furthermore, a self-efficacy test is conducted to determine students with high and low self-efficacy.From the results of this test, 27% of the top ranking are grouped as students with high self-efficacy, while 27% of the lowest ranking are grouped as students with low self-efficacy (Urbina, 2004).Finally, the research sample consisted of 30 students from class A (15 with high self-efficacy and 15 with low self-efficacy) and 32 students from class B (16 with high self-efficacy and 16 with low self-efficacy), so that the total 62 samples.

FINDINGS AND DISCUSSION
This section presents the results of data analysis that has been collected regarding the effect of selfefficacy and gender on the ability to write in English.The results of the study are described based on statistical data that has been processed, including the average value and standard deviation for each group identified based on self-efficacy and gender levels.This finding was then analyzed to identify the pattern and relationship between the variables studied, and compare them with literature and previous research.This discussion aims to provide in-depth insights on how self-efficacy and demographic factors can affect academic performance, especially in the ability to write English, as well as practical implications that can be drawn from the results of this study.

Data Analysis
Table 1 shows the data that the ability to write English in male students who have high selfefficacy consists of 15 students, with an average value of 88.40 and a standard deviation of 5,050.Meanwhile, for male students who have low selfefficacy, which also consists of 15 students, obtained an average value of 70.67 with a standard deviation of 6,608.In female students, the ability to write English for those who have high self-efficacy consist of 16 students, with an average value of 71.09 and standard deviations of 8,043.As for female students with low self-efficacy, which also consists of 16 students, the average value obtained is 78.16 with a standard deviation of 12,366.From this data, it appears that self-efficacy has a significant influence on the ability to write English both in male and female students (cf.Teng & Wang, 2023).Male students with high self-efficacy have an average value that is much higher than those who have low self-efficacy (cf.Guo et al., 2023).On the other hand, female students who have low selfefficacy have an average value higher than those that have high self-efficacy.This difference indicates that self-efficacy does not always affect learning outcomes in similar ways.It depends on other factors such as motivation, environmental support, or academic level of anxiety.A higher standard deviation in groups with low self-efficacy also shows greater variations in the results achieved by students in the group.It shows that low selfefficacy may not only affect the average yield but also the consistency of academic performance.

Normality test Normality of Data A1, A2, B1, and B2
To ascertain whether parametric or non-parametric statistics are used in the following analytic strategy, the dependent and/or covariate variables underwent a normality test.Parametric statistics were used to continue the investigation if the data passed other theoretical distribution tests like the normality test.The normality test findings are displayed in Table 2.It includes the statistical values of the Kolmogorov-Smirnov Z for the following values: A1 = 1,111, A2 = 1,411, B1 = 1,591, and B2 = 0.997, with all sig values more than 0.05.This demonstrates the normal distribution of the data.Kolmogorov-Smirnov Z statistics produced in this study show that all data groups, both based on sex and self-efficacy level, are normal distribution, because the significance value (sig.) is greater than 0.05.It means that the assumption of normality was met.It allowed this study to use parametric statistical techniques in further analysis.When compared to non-parametric statistics, parametric statistics are often stronger and more sensitive to identify differences or correlations between variables.Thus, it is crucial to validate that the data is regularly distributed to guarantee the validity of study findings and appropriate interpretation.
Data Normality A1B1, A1B2, A2B1, and A2B2 Groups A1B1, A1B2, A2B1, and A2B2 refer to pairs of independent variables that need to be checked for normality before moving on to additional statistical analysis.Table 3 presents the results of the test that all groups have a value of P larger than 0.05, indicating that the data is normally distributed.
This study was continued with parametric statistical approaches, which needed normal distribution data to produce valid and trustworthy results, after the data fulfilled the normality assumptions.Therefore, to guarantee that the findings are not influenced by atypical data distribution and that interpretations and conclusions formed from statistical analysis can be relied upon and are valid, this normality test was also an essential stage in the data analysis process.When using statistical analysis techniques, this normal distribution has significant consequences.This study can employ parametric statistical approaches like ANOVA because the data were normally distributed.These techniques are often stronger and yield more accurate estimates than nonparametric methods.This normality assumption ensured that the results of the analysis were more reliable and representative of the population under study.Thus, the study drew more valid conclusions and develop more precise recommendations based on findings.It also strengthened the internal validity of the research because it allows the use of statistical techniques that require certain data distribution to provide valid and appropriate interpretations.

Homogeneity Test
The homogeneity test on the data was carried out using the Levene test at a significance level of 5%.Table 4 shows that the value of Fh = 1,076 and Sig.= 0.366> 0.05.This shows that data comes from homogeneous samples.Thus, the null hypothesis was accepted.It shows that the sample comes from a population with the same variance, or homogeneous population.The result of the Levene test shows that there is no significant difference in the spread of the score of English writing skills among the groups that were compared.This means that these groups have relatively the same variants.If the variance of a large group, meaning that the values in the group vary greatly, ranging from very low to very high.In contrast, if the variance is small, the values in the group tend to gather around the average value.In statistical analysis, especially in the average comparison between groups, the assumption of variance homogeneity is very crucial (Zhou et al., 2023).If the variance between groups is significantly different, the analysis results can be biased and inaccurate.

Hypothesis Testing
After the normality and homogeneity test, and the results show that the samples was generated from a normally distributed population and a homogeneous sample variant, the hypothesis testing was continued using ANOVA.The analysis of Students' writing ability was carried out with two-way ANOVA using the SPSS 2.0 program.The ANOVA test results were then followed by the F test to determine the significance of differences between groups (simple effects).The F test was used to see a group of samples that have higher English writing skills, in terms of gender and self-efficacy perspectives.Table 5 displays the results of data analysis using ANOVA.The proposed research hypothesis can be answered with several explanations.First, in the hypothesis that there is a significant gender influence on the ability to write English, the results of the ANOVA test show the Sig.= 0.007 <0.05 and the Fh = 7,748.This means that the zero hypothesis (H0) is rejected and the alternative hypothesis (H1) is accepted.This indicates that there is a significant gender influence on the students' ability to write English.In other words, there are differences in the ability to write English between male and female students.These findings may be comparable to some studies on gender influence in other settings (Hidayanti & Anggraini, 2023;Shen & Bai, 2024).
Second, in the hypothesis that there is a significant effect of self-efficacy on the students' ability to write English, the results of ANOVA show Sig.= 0.004 <0.05 and Fh = 9,155.This means the null hypothesis (H0) is rejected and the alternative hypothesis (H1) is accepted.This means that selfefficacy has a significant influence on the students' ability to write English.In other words, there are differences in the ability to write English between students with high and low self-efficacy (cf.Sehlström et al., 2023;Shen et al., 2023;Teng & Wang, 2023).
Third, in the hypothesis that there is a significant interaction effect between gender and self-efficacy on the students' ability to write English, the results of ANOVA show Sig.= 0.000 <0.05 and Fh = 49,434, which means the null hypothesis (H0) is rejected and the alternative hypothesis (H1) is accepted.This shows that there is a significant effect of interaction between gender and self-efficacy on the students' ability to write English.Several previous studies in different settings have actually confirmed similar things (see Bai et al., 2022;Chen, 2020;Kutuk et al., 2022).
In the meantime, the Adjusted R Squared value of 0.504 shows that gender and self-efficacy together have an influence of 50.4% on increasing the ability to write English.This indicates that the two variables play an important role in determining the level of English writing ability.The remaining 49.6% may be influenced by other factors that are not examined in this study.These results confirm the importance of paying attention to gender and self-efficacy factors to improve the quality of learning English writing.

Further testing
To find out the extent of the interaction between gender and self-efficacy affect the ability to write English, this study conducted further testing with the Tukey test.Table 6 displays the results.This test aimed to explore significant differences between various groups in this study.Based on the results of the further test, several important findings were found.First, in groups A1B1 and A1B2, an average difference of 17.73 was found.This shows a significant difference in the ability to write English between this group, with a significance value of 0.000 which is far below 0.05.This shows that for group A1, there are significant differences in the ability to write English between sub -groups B1 and B2.
Second, in the ratio between groups of A1B1 and A2B1, the average difference found is 17.31, also with a significance value of 0.000 <0.05.This shows that for Group B1, there are significant differences in the ability to write English between sub -group A1 and A2.
Third, in groups A1B2 and A2B2, the average difference of -7.49 shows that although there are differences, the value is smaller than the previous difference, it remains significant with a significance value of 0.020 <0.05.This indicates that for groups B2, there are significant differences in the ability to write English between sub -group A1 and A2.
Finally, in the ratio between groups of A2B1 and A2B2, an average difference of 7.06, with a significance value of 0.028 <0.05, shows that there is a significant difference in the ability to write English between sub-groups B1 and B2 for group A2.
These results indicate that the interaction between gender and self-efficacy has a significant influence on the ability to write English, with variations in this effect depending on the specific combinations of these variables.This interpretation is important to understand how demographic and psychological factors interact with each other in influencing academic outcomes, especially in the ability to write English.Previously, Getie (2020) had put forward an idea related to this.This study may confirm the validity of that idea.
The interpretation of the continued test results of this Tukey shows that gender and self-efficacy factors not only affect individually, but also interact with each other in determining the ability to write English.In a more general context, Šabić et al. (2022) also put forward a similar idea.This study may help narrow down the research background to a more specific field.The significant differences found between various groups asserted that the combination of self-efficacy and gender levels can produce significant variations in learning outcomes.For example, students with high self-efficacy show better performance than those who have low selfefficacy, regardless of their gender.However, lower results in certain groups show that although selfefficacy is an important factor, there is a possibility that other factors such as teaching methods, environmental support, or previous writing experience also play a role in determining the ability to write.Therefore, this study suggests that the targeted educational intervention to increase selfefficacy, especially in groups with low self-efficacy, can be an effective strategy for improving writing skills.In addition, curriculum development that pays attention to gender and self-efficacy differences can help create a more inclusive and supportive learning environment, so that all students can achieve their maximum potential in writing English.

Further Interpretations of Results
Anova test results show that there is a significant gender influence on the ability to write English.The significance value (Sig.) of 0.007 shows that the difference in writing ability between male and female students is quite significant.In this context, the results of research indicate that female students generally have a higher average value in writing English compared to male students.This can be caused by various factors, including differences in motivation, level of anxiety, and social support that may be higher among women.These findings are in line with some previous studies that show that gender can affect academic achievement in various fields, including language.
In addition, this study also revealed that selfefficacy has a significant influence on the ability to write English.With sig value.equal to 0.004, students with high self-efficacy tend to have better writing skills than those who have low self-efficacy.Self-efficacy, or a person's belief in his own abilities, plays an important role in determining how much effort and perseverance is in completing complex tasks, such as writing in English.Students with high self-efficacy are more likely to overcome difficulties, seek help when needed, and actively find opportunities to improve their skills.
This study also found a significant effect of interaction between gender and self-efficacy on the ability to write English.The results of the analysis show that the combination of gender and selfefficacy level has a different effect on learning outcomes.For example, female students with high self-efficacy show better results compared to men with low self-efficacy, but the same thing also applies to other comparisons involving a combination of these variables.This finding shows that the interaction between demographic and psychological factors is very complex and can affect academic results in unpredictable ways based on only one variable.
To deepen the analysis, Tukey's further test was carried out to explore significant differences between various groups.Further test results show that differences in the ability to write English are not only influenced by gender and self-efficacy individually, but also by interactions between the two.A significant difference between these groups shows that there are considerable variations in the way students process and apply their writing skills, depending on the specific combinations of these factors.
The interpretation of the results of this study confirms that both gender and self-efficacy are important determinants in developing the ability to write English.However, it is also important to consider that other factors such as teaching methods, learning environment, and social support also contribute to learning outcomes.Therefore, educational interventions targeted to increase selfefficacy, especially among students with low selfefficacy levels, can be an effective strategy for improving writing skills.In addition, curriculum development that considers gender and self-efficacy differences can help create a more inclusive and supportive learning environment, so that all students can achieve their maximum potential in writing English.By understanding these dynamics, educators can develop more effective and adaptive English writing strategies, and create a learning environment that supports the academic and personal growth of all students.

CONCLUSION
This study has revealed that there is a significant influence of gender factors on students' English writing skills in the central Indonesian region.A comprehensive analysis of the results of the study shows that there are significant differences in writing skills between male and female students.This difference reflects a significant gender disparity in the development of English writing skills in an academic context.This finding emphasizes the importance of considering gender as an important factor in developing English writing competence among students.
The results of the study further show that there is a significant disparity in the English writing skills demonstrated by male and female students.An important indicator of this disparity is seen from the examination of the average scores associated with each gender.Specifically, the average score indicating the English writing skills of male students is [the actual average score of male students].Meanwhile, the average score of English writing skills for female students is [the actual average score of female students].This difference in scores emphasizes the significant differences that emerge in the context of the influence of gender on English writing skills, which are worthy of further consideration and research in academic discourse.
In addition, this study also revealed that selfefficacy has a significant influence on students' English writing skills.With a significance value (Sig.) of 0.004, this study shows that students with high self-efficacy tend to have better writing skills compared to those with low self-efficacy.Selfefficacy, or a person's belief in their own abilities, has been shown to be an important factor in determining how much effort and perseverance students expend in completing complex tasks such as writing in English.There was also a correlation between gender and self-efficacy on Students' English writing skills.The results show that the combination of gender and self-efficacy levels provided different effects on learning outcomes.For example, female students with high self-efficacy had better results compared to male students with low self-efficacy.However, the same effect was also found in other comparisons involving a combination of these variables.These findings suggest that the interaction between demographic and psychological factors is very complex and can affect academic outcomes in ways that cannot be predicted based on just one variable.
The correlation between gender and selfefficacy creates a unique landscape of English writing skills was also found.These two variables brought dynamics in the development of students' writing skills.They are seen through the analysis of mean scores that the combined impact of gender influences and dynamics are related to self-efficacy.Specifically, the mean scores of English writing skills reflect the combined impact of gender and self-efficacy factors.
Overall, this study suggests that gender and self-efficacy factors play an important role in determining the level of students' English writing skills.These results confirm that educational interventions aimed at increasing self-efficacy, especially among students with low levels of selfefficacy, can be an effective strategy to improve writing skills.In addition, developing a curriculum that considers gender differences and self-efficacy can help create a more inclusive and supportive learning environment, so that all students can reach their maximum potential in English writing.By understanding these dynamics, educators can develop more effective and adaptive teaching strategies and create learning environments that support the academic and personal growth of all students.

Table 4
Homogeneity Test: Levene's Test of Equality of Error Variance

Table 5
Research Hypothesis Test: Tests of Between-Subjects Effects

Table 6
Further Test Table: Multiple Comparisons