7/28/2019 115829013 Defining Resarch Question
1/61
Defining the Research QuestionPrepared bY : Smirkou Ahmed,
Hammani Mohamed, Idabdallah
Abdelhadi
7/28/2019 115829013 Defining Resarch Question
2/61
7/28/2019 115829013 Defining Resarch Question
3/61
outline Sources of research questions
Scope of research
Feasibility of research Stating research questions and hypotheses
Collecting research evidence
Internal and external validity of studies
7/28/2019 115829013 Defining Resarch Question
4/61
Definition of the concept Research Research has been defined as a systematic approach to
searching for answers to questions. (9)
7/28/2019 115829013 Defining Resarch Question
5/61
7/28/2019 115829013 Defining Resarch Question
6/61
1) Sources of Research QuestionsA) From the teachers themselves
o for them, research was something that people did at
universities or labs. They described research as havinggreat theoretical scope,with numbers and symbolsthat made it difficult to understand. (9)
7/28/2019 115829013 Defining Resarch Question
7/61
A List of Questions1. Do teachers have questions about teaching?
2. Do they wonder about the use of teaching materials,
techniques, etc.?3. Do they share such questions with their students or
with other teachers?
4. Why some teachers believe that research should be
conducted elsewhere?
7/28/2019 115829013 Defining Resarch Question
8/61
Research Interests Top Down instructions:
Devorced from true curiosity
The results are probably predictable It is like being asked to write a composition with an
open-choice topic:
Lack of directions
Much time is spent complaining or worrying about the wayyou can please the person who gave directions
Supervising Faculty
7/28/2019 115829013 Defining Resarch Question
9/61
Questions are our own Questions we truly want to investigate
Questions must be important to us
Our questions are shaped by our experience Our teachers,colleagues, students, and the reading we
do guide us towards the important issues in our fields
7/28/2019 115829013 Defining Resarch Question
10/61
B)Another source of research:Journals of ourfield
They show which questions are especiallyinteresting at a particular time
The wide range of possibilities for research
They offer specific ideas for further research
Some researchs major areas:
Learner /teacher
Characteristics
ClassroomResearch
Languageanalysis
7/28/2019 115829013 Defining Resarch Question
11/61
Picking up particular research
topics Reasons:
We research what we like or what appeals to us
We research what we think we already know How to define what interests you:
1. Use a research journal
2. Form a study group: find critical friends/ friendlyenemies
7/28/2019 115829013 Defining Resarch Question
12/61
Practice1.1 Start a research journal:
Write your notes and questions in the journal for one week
Compare the research interests of the study group members
Report on a range of interests
Pick up a certain article or a research topic from yourrecent journal:
7/28/2019 115829013 Defining Resarch Question
13/61
7/28/2019 115829013 Defining Resarch Question
14/61
In addition to Personal Interest, researchquetions have other characteristics:
to generate new information
to conform old information in new ways
Thats why a review of the literature on theresearch topic must be done
This helps you locate your research area. Also it helps you not to select the topic if you
thought that it had already been sufficientlyaddressed.
7/28/2019 115829013 Defining Resarch Question
15/61
An Example: If you are to carry out your research on how students
can perform basic SpeechActs1) First, you need to undertake a review of previous speech
act: the university library might helpo Look for key words and key authors : in the case of
Speech Act, you might find terms like directive,assertive, commissive or such authors asAustin, Searle,Gumperz.
2) The topic is still too broad3) Narrow down the scope of the search
4) Ask yourself such questions: Which speech act do I want to investigate? What type of learners/ students am I going to talk about? Which area am I going to focus on? What kind of situations should be investigated?
7/28/2019 115829013 Defining Resarch Question
16/61
The process of narrowing the
scope of research1) Via key words:
Speech Acts
Directive, commissive, etc.2) Via key setences:
Investigate how well second langauge learners performspeech
acts
Investigate MoroccanESL students ability to recognizecomplaint behaviour appropriate in an accademic universitysetting
7/28/2019 115829013 Defining Resarch Question
17/61
Even a fairly narrow question may need to be more
carefully defined because previous researchers may
already have done it.
Use operational definitions to subcategorize or narrow
the concept:
Motivation
Intrinsic extrinsic instrumental integrative
7/28/2019 115829013 Defining Resarch Question
18/61
Operational definitions give a tried and true definition and anaccepted method for observing or assessing it. (15)
Sometimes there are no accepted definitions or no agreement as towhat the terms mean as the case of abstract theoretical concepts such
as the terms motivation, acquisition, bilingualism. (15) Terms must be operationally defined: an operational definition is aclear statement of how you judge or identify a term in your research. (15)
Ex:Acquisition in my study means andwill be measured or observed by..
Good operational definitions can be drawn from the existing literature Sometimes research is difficult to carry out because operational definitions cannot be found that will
satisfy the researcher. Sometimes no operational definitions exist in the literature and the researcher must define terms.
Thats why the attempt to establish concepts is an important area of research.
7/28/2019 115829013 Defining Resarch Question
19/61
The operational definition is
important for three reasons:1. You will need to be absolutely consistent throughout the
research processin your definition.
2. It is important for consumers of your research so thatthey do not misinterpret your findings.
3. It is important to the research community that yourstudy be replicable.
A Similar project but a different definition mayresult in different findings.
7/28/2019 115829013 Defining Resarch Question
20/61
Balance between scope and interest
We may lose the interest in the topic because it nolonger addresses the larger question
Try to maintain the original research interest bycarrying out a number of studies with limitedscope
7/28/2019 115829013 Defining Resarch Question
21/61
7/28/2019 115829013 Defining Resarch Question
22/61
Research questions should:1. Interest us
2. Promise new information or conform oldinformation in new ways
3. have reasonable scope
4. have key terms that are clearly defined andoperationalized. (17)
7/28/2019 115829013 Defining Resarch Question
23/61
Before making a further step or stating the questionsin a more formal way, we need to consider whether ornot the research is feasible.
The factors that affect the feasibility ofresearch: know the time span it will take
do you have that amount of time to spend?( when the topic is very broad, it may take a lifetime to be
investigated)
7/28/2019 115829013 Defining Resarch Question
24/61
Longitudinal studies: it follows an individual orgroup over a period of time.
Cross-sectional approach: data are gathered from
different groups of learners of different ages ordifferent levels of proficiency.
Longitudinal study of Luisa
0:9 1:0 1:3 1:6 1:9 2:0
Cross-Sectional Study of 30 children
5 at 0:9 5 at 1:0 5 at 1:3 5 at 1:6 5 at 1:9 5 at2:0
7/28/2019 115829013 Defining Resarch Question
25/61
In Cross-sectional approach: data could becollected at one time rather than spread out overtwo years.
This makes the study more feasible.
However, age may not be a very satisfactory way toequate stages.
The researcher must be able to locate appropriatelearners and have time to collect the data fromeach of them.
7/28/2019 115829013 Defining Resarch Question
26/61
By quantity and quality of access: For example assume thatthe government is interested in finding out the extent ofbilingualism in your country. One way they could do this isby including questions on the next United States census
questionnaire. First, the qualityof the sample is likely to be biased given
the existence of undocumented aliens, who may or may notwant to be part of the census count.
Second, quantity(the number of questions that could beallocated to this issue on the questionnaire) would beseverly constrained
Third, the cost, if bilingual census takers must be found toconduct census taking,might not make the project feasible
7/28/2019 115829013 Defining Resarch Question
27/61
Feasibility may be determined Bytime
By access: you may have great difficulty in gainingaccess to the classroom. Many schools andteachers require a statement that the research willnot disrupt regular instruction. Restrictions maybe placed on the data-gathering procedure by theschool.
7/28/2019 115829013 Defining Resarch Question
28/61
The cost of research in planning a
project Prepare a reasonble budget
Do you need tape recorders and tapes?
If videotaped data are required for your study, are
videocamera and tapes available? Can you operate the videocamera and observe a class
at the same time, or must you hire a camera?
Do you need paper supplies, travel to and from a
school,photocopies? List the procedures needed to gain access to the
classroom
Try to make a complete list of everything you need
7/28/2019 115829013 Defining Resarch Question
29/61
7/28/2019 115829013 Defining Resarch Question
30/61
Stating Research Questions and
HypothesesAfter the scope and feasibility of research questions, it
is time to consider how these questions can be clearlystated:
Imagine you want to describe the bilingual languagedevelopment of the child, Luisa.
Your basic research question:
Can I describe the bilingual languagedevelopment of Luisa?
Can I describe the first 50 Spanish and first 50English words acquired by Luisa?
7/28/2019 115829013 Defining Resarch Question
31/61
What do you expect in the data? Do you expect to see a change in meanings over time?
Do you expect to see parallel words in the twolangauges? Etc.
7/28/2019 115829013 Defining Resarch Question
32/61
We often begin a research with questions like:
Can I describe X ? Rather than Why do I want todescribe X ?
Can I describe questions are often difficult, for weseldom know how we will go about this descriptivetask until we begin to examine the data.
7/28/2019 115829013 Defining Resarch Question
33/61
Example Consider the work that is done in needs-assessment
research: One basic question that pops up:
Can I describe how X perceive their needs regardinginstruction ( or whatever)?You should expect what to find out:If you involve parents in the instruction process, do you
expect all parents to respond in the same way? If not, what
factors might help explain differences in their responses?...Once we start thinking about what might happen inthe data, we begin to think about how to explain thefindings.
7/28/2019 115829013 Defining Resarch Question
34/61
Practice1) Select one question from your research journal.
Explain why it is an interesting question to the field.
2) Frame the general research question for the study.
3) Consider the possible outcomes of the study. Whatfactors may enter into the research that mightinfluence the outcomes.
7/28/2019 115829013 Defining Resarch Question
35/61
It is time to: Read
Observe
Think about the project very carefully Look at your research journal and revise your research
question many times and think about manyaternatives and the possible outcomes.
Dont plunge into a project without taking time to letyour minds sort through many possibilities
7/28/2019 115829013 Defining Resarch Question
36/61
HYPOTHESES When we have questions about something, we reallydont
know the answers for sure. This doesnt mean that we have no idea about what those
answers might be or where to look for them. Our hunches about answers may come from reviewing the
literature on the toip, from talking with colleagues, orfrom observing classrooms.
These hunches about answers,when written in a formalway, are called hypotheses.
We carry out our research to see if the hypotheses aresupported or not.
Dont forget that research is a way to prove that an answeris right or wrong.
7/28/2019 115829013 Defining Resarch Question
37/61
You may state the alternative hypothesis in adirectional form ( positive or negative):
Ex: There is a positive/negative relationship betweenL2 proficiency and placement on the spellingcontiuum.
7/28/2019 115829013 Defining Resarch Question
38/61
It will seldom be possible to prove that your answer is theright one.
A hypothesis is a statement of possible outcome ofresearch.
A hypothesis may be stated as:
1) a nullhypothesis. Ex: There is no order of acquisition of
English spelling patterns: in null hypothesis, there is norelationship between L2 proficiency and placement onthe spelling continnum.
2) an alternativehypothesis. Ex: There is an order of
acquisition of English spelling patterns: in alternativehypothesis, there is a relationship between L2 proficiencyand placement on the spelling continnum.
7/28/2019 115829013 Defining Resarch Question
39/61
The null hypothesis is tested rather than a
directional alternative hypothesis
It is easier to find evidence that supports a directionalhypothesis than it is to reject a null hypothesis.
Different statistics will be used on the distinction ofwhether the hypothesis is directional.
Sometimes you need to write more than onehypothesis to cover the research question
7/28/2019 115829013 Defining Resarch Question
40/61
-1 Collecting research evidence.-2 Internal and external validity..
7/28/2019 115829013 Defining Resarch Question
41/61
7/28/2019 115829013 Defining Resarch Question
42/61
DEFINITION
Scientific evidence: evidence which
serves to either support or countera scientifc theoryor hypothesis.
7/28/2019 115829013 Defining Resarch Question
43/61
What should be done?When planning a research one should
consider the kinds of evidence that are
needed.Your findings will allow you to SUPPORT or
REJECT your hypotheses.
Data collection: how to find the mostEFFICIENT way to collect data. Theresearch question(s) determine(s) your data
collection method.
7/28/2019 115829013 Defining Resarch Question
44/61
Efficiency: depends on the amount of time onehas to carry out the project in.
Ample time allows you to think of differentways to gather evidence.
- Good BUT: 1/ Time is mostly restricted. 2/toostrong.
Ex: Suppose you want to investigate the typesof spelling errors made by men and womenfrom different first language groups:
7/28/2019 115829013 Defining Resarch Question
45/61
In this case There are a series of questions to beanswered like:
- How might you OPERATIONALLY
define advanced so that you are sure the samples comefrom the appropriate group?
- how many compositions will youneed?
- Will words representing all thespelling patterns actually occur in all compositions?
- HOW LONG will it be before all thesamples arrive?
7/28/2019 115829013 Defining Resarch Question
46/61
Hence EFFICIENCY isnt
everything If the method you use isdull or frightening or TAKES TOOLONG, it is unlikely that your subjects
will be motivated to perform as well asthey might.
7/28/2019 115829013 Defining Resarch Question
47/61
The data collection method should notonly motivate subjects to participate,but should allow them to give their best
possible performance.
7/28/2019 115829013 Defining Resarch Question
48/61
Questionnaires Time requirements and task boredom can discourage
people from responding.
To ensure a better way of response and return ofquestionnaires, its important to consider exactly whatinformation must be obtained from the respondentand what information can EASILYbe gathered fromother sources.
7/28/2019 115829013 Defining Resarch Question
49/61
Internal and external validity Cambel and stanely(1963), note that internal validty
has to do with the interpretation of the researchfindings within the study iteself. Whereas external
validty, has to do with the extension of the researchfindings beyond the confinments of the study.
7/28/2019 115829013 Defining Resarch Question
50/61
Threats to internal validity
Subject selection : When conducting a
reserach, the subjects or your sample grouptend to bring with them certaincharacteristics, some learnt and someinherent. Examples inculde : sex, attitude,personality,mental ability etc).
7/28/2019 115829013 Defining Resarch Question
51/61
During our review and assessment of subject selection,we should consider the following issues:
- The purposes of the research.- The setting in which the research would beconducted.
- The Population.
7/28/2019 115829013 Defining Resarch Question
52/61
Maturation : In any research, there is always thethreat that the subjects may change in the course ofthe experiment due to the passage of time. Forexample BECAUSE OF FATIGUE.
7/28/2019 115829013 Defining Resarch Question
53/61
History : Refers to the outisde events that mayinfluence subjects in the course of an experiment orbetween repeated measures of the dependent variable.
Suppose that the dependent varibale is measuredtwice for a group of subjects, once at time or place Aand later at time or place B, in this case theindependent varibale changes and therefore the results
are not valid.
7/28/2019 115829013 Defining Resarch Question
54/61
Instrumentation : refers to theobjectivity and validity of the research
measurments. Data that isnonobjective threatens a studysinternal validity.
Ex : questionnaire with grammarquestions to test fluency notvalid.
7/28/2019 115829013 Defining Resarch Question
55/61
Task directions : Refers to howvalid are the instructions of theinstrument used in the reserach.Long , hard to understand, time
demanding etc.
7/28/2019 115829013 Defining Resarch Question
56/61
Adequate Data Base : Here, if you are to measureone thing with different items, these items should becarefully arranged. That is , there should be some
sort ofconsitency; The items should be close to eachother in terms of expectation on the part of thesubject so that it is easy for them to fill it honestlyand comfortably. This is called : FORMING A SET
RESPONSE. To be valid, the data gathering procedure should
allow us to tap the true abilities of the Subjects.
7/28/2019 115829013 Defining Resarch Question
57/61
Testing effects consist of either reactivityas a result of testing or practice/learning
from exposure to repeated testing.Longitudinal studies which requireparticipants to take certain tests on anumber of occasions are subject to thisthreat to internal validity.
7/28/2019 115829013 Defining Resarch Question
58/61
External validity
If a study deos not have internal valdity, it cant havethe external one. we cannot generalize.
How representative the data are for the group(s) towhich we hope to generalize?
7/28/2019 115829013 Defining Resarch Question
59/61
Threats to external validity Sample selection: the selection SHOULD NOT be by convenience sampling that is, they happened to be the
group that was available.
EX: if we have specified the population as ESL universitystudents elsewhere, then our sample Ss must be selected toREPRESENT ESL students everywhere.
Hence: The group should be slected RANDOMLY.
HOWEVER: There are problems with this method; forexamlple: There may be more men that women , moreimmigrant that foreign students, more students from thesciences that humanities etc. ONE WAY TO SOLVE THISIS THROUGh a STRATIFIED RANDOM SAMPLE.
7/28/2019 115829013 Defining Resarch Question
60/61
Finally:
In order to avoid threats to external validity, a detaileddescription of the population to which you willgeneralize must be given, Ss must be randomly slectedusing that description, and threats to internal validitymust be met. Otherwise, no generalization can bemade.
7/28/2019 115829013 Defining Resarch Question
61/61
Thank you for your
attention