Methodology semestre 3

SEMESTRE 3: RESEARCH METHODOLOGY IN LINGUISTICS Semestre 3 RESEARCH METHODOLOGY IN LINGUISTICS

Four Lessons 1. Linguistic research methodology 2. Research methods 3. Research tools/ data collection techniques or strategies 4. Data analysis

Lesson 1 1. RESEARCH METHODOLOGY IN LINGUISTICS Origin of Method: is the Greek word “ methodos ” which consists of two parts: “ meta ”: “after”, and “ hodos ” : “road” Methods = procedures = techniques = approaches = ways Methodology is to analyse research methods

WHAT IS LINGUISTIC RESEARCH METHODOLOGY ? The analysis of research methods used by linguists to collect data related to a topic in linguistics .

Writing The Research Proposal INTRODUCTION 1. Review of Literature 2. Statement of the Problem 3. Aims of the Study 4. Hypothesis 5. Research Methodology and Design 5.1 Choice of the Method 5.2 Population of the Study 5.3 Data Gathering Tools 6. Structure of the Dissertation CONCLUSION BIBLIOGRAPHY

Population of the study = Sample = Participants = Informants = Subjects =

Lesson two 2. RESEARCH METHODS QUANTITATIVE METHODS QUALITATIVE METHODS

Hypothesis: If A happens, B would occur. A : the assumed cause (stimulus) B : the assumed effect (response) Cause and effect relationship/ causal relationship.

Example of a hypothesis If students use their metacognitive strategies, they would become autonomous learners . Use of metacognitive strategies: the cause: independent variable. I.V Autonomy: the effect: dependent variable. D.V

The null hypothesis: H 0 What people believe is true. But the researcher thinks it’s false. Example: If students work in groups (co-operative learning), their writing proficiency would be high.

The alternative hypothesis H 1 What the researcher believes is true. It is his/her research hypothesis . Eg. If students work individually, their writing proficiency would increase. Because s/he thinks that: if students work in groups, their writing proficiency would decrease.

QUALITATIVE METHODS Descriptive method Historical method

CORRELATIONAL METHOD Correlation = the relationship between two variables. When do we follow this method? When both variables are countable/ measurable . Eg. Intelligence and academic achievement.

DESCRIPTIVE METHOD When? 1. When one variable is uncountable . Eg. Attention in the classroom and academic achievement. 2. When the two/both variables are uncountable. Eg. CALL and motivation . 3. When we have one variable : students’ lack of vocabulary.

EXPERIMENTAL METHOD 1. When something doesn’t exist. 2. When we can realize it: feasibility/practicability of research. 3. When we cannot count the correlation coefficient. Eg. Cooperative learning and high writing proficiency.

« r »: coefficient of correlation If r is near +1 or -1 the correlation is high. ( 0.95 ; 0.8 ; -0.7 ...) If r = +1 or -1 the correlation is strong/perfect/high. If r is between “ +/- 0.25 ” and “ +/- 0.75 ” it is a moderate correlation. If r is near 0 the correlation is weak If r = 0 there is no correlation: no relationship between variables

If r is positive (marked by + ) this means that if “ x ” increases ,“ y ” also increases .( linear relationship) if “ r ” is negative (marked by - ) this indicates that if “ x ” increases , “ y ” decreases . X : 1 st variable: cause Y : 2 nd variable: effect

MIXED METHODS The Experimental-correlational method (two methods) When? When we can conduct an experiment + we can count correlation coefficient. E.g. the influence of using collocations on students’ writing proficiency. Experiment: teaching collocations

Experimental design At least 2 groups : 1.The experimental group 2.The control group

1.The experimental group: It receives the experiment = treatment = intervention Example: If we teach students grammar, their writing proficiency would raise. Teaching grammar is the experiment.

2.The control group: it doesn’t receive the experiment/treatment. It’s a comparison group ( we don’t teach them grammar ) Why do we need this group? to compare the results of the experimental group to those of the control group in order to see whether the experiment was effective. If the writing proficiency of the experimental group has raised.the experiment was effective (successful)

Sampling 1. THE PROBABILITY/ RANDOM SAMPLE: generalization is possible 2. THE NON-PROBABILITY/ PURPOSIVE SAMPLE: generalization is impossible Sampling sample

1. THE PROBABILITY/ RANDOM SAMPLE

2. THE NON-PROBABILITY/ PURPOSIVE SAMPLE

CSR: C ase S tudy R esearch a single case a group of cases/ multiple cases CSR DESIGN a single case: a student, a school a group of cases/ multiple cases: st1+st2+st3 or sc1+sc2+sc3

Generalization in case studies It depends on: 1.The sample: random or not random 2. The method: quantitative, qualitative, qualitative + quantitative

Types of CSR: Robert K. Yin ( Case Study Research: Design and methods, 1993 ) 1.EXPLORATORY: before research 2.EXPLANATORY: only the case study, no research after it. 3.DESCRIPTIVE: before research, there is a descriptive theory

Types of CSR: Robert E. Stake ( The Art of Case Study Research: 1995) 1. INTRINSIC 2.INSTRUMENTAL 3.COLLECTIVE (DESIGN)

YIN CSR METHODOLOGY Conclusion Recommendations Implications

CSR METHODOLOGY 1. research question(s) 2. case(s)+ way of data collection and analysis 3. preparation to collect data 4. data collection in field 5. data analysis 6. writing the report

CSR Sources of evidence 1. Documentation

THANK YOU FOR PAYING ATTENTION TO THE LESSON

Lesson Three 3. Research tools: data collection techniques/ strategies

We have many research tools. We’re going to deal with: 1. The questionnaire 2. The interview 3. Observation 4. The pretest and the post -test

1. The questionnaire To operationalize the questionnaire = to make it structured as much as possible

Structured = there is wording / written form Example: choice 1: 15 students Question 1 choice 2: 5 students choice 3: 10 students scoring / coding : numbers

Structured way of data collection Unstructured way of data collection There is wording (words) Written form Data type is quantitative / numerical: there are numbers, scores, measurements,percentages ( ℅)… No wording (words) No written form Data type is qualitative / word-based / narrative: there are words .

Interview types: according to structure

2. The semi-structured interview No questions, no wording, just topics of questions. Example: Question 1: question topic: definition of motivation. Question 2: question topic: types of motivation. So, no wording for each question. i.e. the question is not written : different forms of questions (different wording) for different informants but the same topic.

Question 1: definition of motivation Informant 1: Do u know the definition of motivation? Informant 2: What is motivation? Informant 3: How could you define motivation?

Interview types: according to the number of interviewees T w o t y p e s

Interview design = The nature of the questions: 1. Open and/or closed questions. 2. Direct and/or indirect questions: Specific and/or non-specific (general) questions

3. Observation To get ‘live’ data from ‘live’ situations.

According to Patton observation is “ to look at what is taking place in situation rather than at second hand” (cited in Cohen, L , Manion, L and Morrison, K. 2000: 305).

Observation types: according to structure

1. Structured/standardized observation: topic + hypothesis (hypothesis-testing) numerical data 2. Semi-structured observation: topic but no hypothesis (hypothesis-generating) 3. Unstructured observation: no topic, no hypothesis (hypothesis-generating)

Observation chart / schedule Structured/standardized observation: topic (students’ interaction in the classroom) + hypothesis (if students’ interact with each other, their oral performence would improve)

Degrees of Participant/ naturalistic observation

1. The complete participant complete participation in daily activities. He lives within a group like the spy. covert research: secret. Negative point: s/he could be influenced by the group. So s/he may stop to act a researcher.

2. The participant-as-observer: s/he participates in the group but they know s/he is a researcher: overt research

3. The observer-as-participant s/he doesn’t participate: marginal observer : overt research (not covert)

4. The complete observer s/he doesn’t participate. s/he observes secretly. Covert research Example: a teacher who observes students in the classroom.

4. The pretest and the posttest Testing = measuring . What? an aptitude, an ability, a skill, knowledge.. Examples: - verbal aptitude tests: words - I.Q. tests. I ntelligence Q uotient. ( Q uotient = ratio).

I.Q. = MA / CA × 100. MA: Mental Age CA: Chronological Age

1. The pretest: before the intervention /experiment/ treatment. 2. The posttest: after the intervention /experiment/ treatment.

The pretest To see if the level of students in the experimental group and the control group is equal

The same pretest for both groups

The posttest To see if the level of students in the experimental group and the control is not equal : a change in the experimental group due to the experiment/intervention/the treatment.

The same posttest for both groups

Experiment methodology (steps) 1. Research problem/question 2. Hypothesis 3. Sample choice (random: experimental/not random: quasi-experimental) 4.Experimental design: two groups: experimental group + control group

5. Students’questionnaire and/or teachers’ questionnaire (+ Interview if possible) 6. Pretest + experiment + posttest 7. Data analysis 8.Conclusion: confirming or rejecting the hypothesis

The Solomon four-group design Randomiz- ed groups Pre-test experiment Post-test Group 1 Group 2 Group 3 Group 4

The Solomon four-group design We add two control groups. Why? to eliminate the effect of the pretest: 1. on the posttest: group 3: it’ s the experiment that affected the posttest. 2. on the posttest: group 4: and it is not the experiment that affected the posttest.

Lesson 4 4. Data analysis: 4.1 Analysis of Quantitative data 4.2 Analysis of Qualitative data

Data-collection strategies 1. Longitudinal strategies: individual change over time. Two samples. Example: The influence of teaching oral expression on students’ speaking: two samples: First sample: first year students (2011-2012) Second sample: second year students (the same students: 2012-2013)

2. Cross-sectional strategies: differences between groups at one point in time. one sample. Students’ pronunciation in second year: difference between two groups.

Analysis of quantitative data Example: Analysis of data driven from students’ questionnaire A questionnaire has been administered to collect information from first year students of English who have been allocated randomly to two groups: an experimental group that will receive the treatment and a control one which stands as a means of comparison to see whether the treatment has come to any changes. The aim behind this questionnaire is to collect data about students’ level in writing and their knowledge of collocations.

Analysis of Results and Findings The answers collected from students’ questionnaire have been counted and organized in tables in order to quantify the results which are presented below.

3- Students' choice to study English at the university The experimental group The control group number percentage number percentage Yes 17 70.83 ℅ 16 66.66 ℅ No 7 29.16 ℅ 8 33.33 ℅ Total 24 100 ℅ 24 100 ℅

COUNTING PERCENTAGES NUMBER × 100 / 24 Example: 17 × 100 / 24 = 70.83 ℅

3- Students' choice to study English at the university

Methodology semestre 3

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Methodology semestre 3 (20)

Recently uploaded (20)

Methodology semestre 3