Research starts with delineating the research problem, i.e. what we want to solve and what questions we want to answer.
It is not only forming the problem but also its selection which presupposes experience and overview in the given area. The research problem should ripe and appear gradually. The final form of the problem must be preceded by studying literature and mapping what has already been found out and how.
It is advisable to form the research problem as a question. It helps to decide about the way of research. The whole research then presents an answer to the question.
After the problem has been delineated, it is necessary to define basic terms. Dictionaries and encyclopedias are used to define terms (e.g. for pedagogy it is Průcha, Walterová, Mareš: Pedagogical Dictionary, 2001). Terms must be defined in such a way so that they are allowed to be found and measured. They are operation term definitions such family, quality of life, school maturity or indirect learning strategies. For instance, family is a very broad term; everyone imagines a different meaning. It is necessary to define the term in such a way to be able to work with it.
THE MOST FREQUENT MISTAKES WHEN FORMING A RESEARCH PROBLEM
The author specified the topic (area), not the research problem
The problem specification is too broad. What has been specified is only the area, not what we want to explore and reach.
Example of a mistake: “Problematic teaching of biology”
It is not obvious what is going to be explored: whether the characteristics of the teacher (age, gender, experience) can affect the use of problematic teaching; or the procedures of the teacher in problematic teaching. Or how the students accept this teaching method etc.
The research problem must be formed precisely. The title of the paper can be broad; still, it must point out what is going to be explored. It must also reflect the contents of the paper. A broader title can be more comfortable for the reader’s orientation. It will be specified in greater detail anyway.
The research problem is not of a value or does not make sense
Exploring the problem does not deepen the knowledge on the issue and does not bring anything new.
Example of a mistake: “Does home reading help to develop a student?”
The answer is clear – yes, it is not necessary to prove this. To make it possible to further explore the problem, it must be divided into smaller sub-questions (conditions under which it is performed, the aspects of personality which should be developed…). It must also be specified what kind of literature is meant.
The research problem is trivial, too simple
The answer to the problem is either “yes” or “no”. The problem must be divided into several sub-questions – problems usually cannot be solved with easy answers like “yes” or “no”.
Example of a mistake: “Are elementary school teachers creative?”
As the answer is either “yes” or “no”, it does not advance our knowledge. It is necessary to answer many questions: How is creativity in teaching of a specific teacher manifested? What is its influence on students? In what situations can teaching be creative and when is it routine? Etc.
TYPES OF RESEARCH PROBLEMS
It is good to know 3 basic types of research problems. It will assist in forming them. The type of a problem also determines the use of research and statistical methods.
descriptive
A question of the following type is asked: “What is it like?” Then, we find out about and describe a situation, condition or occurrence of a phenomenon.
Methods: questionnaire, observation, interview, scaling…
Example: What is the proportion of individual teaching activities of teachers of civics at elementary schools?
=> frequency, duration of activity, the proportion of both (observation)
Example: What is the approach of parents whose child is about to fail towards to the teacher and school management?
=> scale (assessing different aspects of the teacher’s professionalism – qualification, teaching style, evaluation…)
Diagnostic-evaluative descriptive problem (for practice, it does not develop theory)
Example: What study results did students of alternative schools reach?
=> didactic tests
Example: What percentage of children from minority groups attends schools in this region?
relative
Two or more phenomena or agents are related. We ask if there is a relationship among the explored phenomena and how close the relationship is. With this kind of problem, it cannot be stated exactly which phenomenon is the cause of which. For instance, questionnaire cannot find out whether a certain opinion was caused by reading a magazine, or whether the student’s opinion led to reading a certain magazine. The cause can be found only in things which are absolutely clear – when we know it from the outside but not because of data. Causal relationships can be verified by experiments.
Statistics: correlation, factor analysis…
Example: What is the relationship between the self-conception of a student and his/her results in key subjects at a secondary school?
Example: What is the dependency between the result of an entrance test and the final result of studies at university?
Example: What is the relationship between child drug addiction and the socio-economic status of the family?
causal
It finds out about the cause which led to a certain consequence; about causal relationships.
Experiment with two groups of persons is carried out.
Statistics – finding the significance of differences: Student’s t-test, chi square…
Example: Is nondirective educational style more effective in creating positive opinions of students on a teacher than a directive one?
Example: What is the influence of a change in break routine at a primary school on students’ ability to work?
SUMMARY:
Most research topics allow us to form research problems of all three types.
Example: The teacher’s use of praise.
Descriptive: What kind of praise do teachers use?
Relative: What is the relationship between the kind of praise used by a teacher and the students’ performance?
Causal: What is the effectiveness of a postponed praise on the students’ performance in comparison with immediate praise?
Scientific hypotheses may be formed only for relative and causal research problems. Why? Since hypothesis is defined as a statement about a relationship between two variables.
The easiest one (time, means) is descriptive research problem. Then it is followed by relative and only then by causal.
On the other hand, causal problem is of the greatest value in pedagogical theory. It is then followed by relative and only then by descriptive problem.
Variables
A variable is a phenomenon, quality, condition or agent which is being explored; e.g. age, knowledge, IQ, the length of practice or behavior. It is a unit of exploration which can get different values which must be defined. For instance, gender is of two values (male – female); for our purposes, marital statues can be only of two values (single – married), at a different time, however, it can be of four values (divorced, widowed), still at a different place even more (separated).
Variables must be turned into operations – defined operatively – so that they could be measured, found and observed. For instance, the ability to speak a foreign language can be defined as a result of a known test, interest in history as a number of questions asked by a student in a history lesson, the number of books on history read by a student or a membership in a history interest club. Therefore, we do not say “what it is” but how we are going to observe it.
It is advisable to differentiate between two main kinds of variables; it is important also for statistical data assessment. With measurable variables, it is possible to define number, degree of a certain phenomenon or quality. Variables get values within a certain range (between – worse; more – less; sooner – later). Example: knowledge, number of mistakes in spelling, favorability of a subject on a scale, a neurotic child according to Eysenck questionnaire.
Categorical variables cannot be quantified; they only can be divided into classes, categories. They can be dichotomy variables (sex: female – male; test result: passed – failed) or not (marital status, grade: 1-9, school subject, method, parents’ education).
Independent variable – it is a cause of a change in the other variable.
Dependent variable – it is the one which changes being affected by a different variable. It is dependent on the one which affects it (independent variable). Example: students’ results (they change under the influence of teaching style).
Hypotheses
The research problem forms the basic focus of the research; however, it does not communicate sufficient information to direct the research. Therefore, hypotheses are needed as they are more specific. Hypotheses divide the research problem into smaller parts, they control the whole structure of the research, and they are validated or invalidated.
Example: Formulating research problem
What is the influence of a teaching style on the student’s performance?
Formulating hypotheses:
H1 Elementary school students’ approach to a subject is more positive when it is taught by a teacher with the non-directive teaching style than by a teacher with the directive teaching style.
H2 Elementary school students have lesser knowledge in a subject which is taught by a teacher with the non-directive teaching style than in a subject taught by a teacher with the directive style.
Hypothesis is a scientific presupposition; it is drawn from a theory (which requires a lot of reading and thinking). It is thus no any presupposition. Rarely is it based on personal experience and general knowledge; this happens only when nothing has been known in such matter so far and the paper presents the first action concerning the issue
Hypotheses broaden our knowledge and learning – they test parts of a theory empirically. On the basis of new findings, theories can be broadened or modified.
Research is time-, effort- and money-consuming. Therefore, it is important to define such hypotheses which are worth the effort, i.e. worthwhile hypotheses. The most valuable are creative hypotheses which are able to advance our knowledge.
Hypothesis controls the research; therefore, it is not possible to start with collecting data and form hypotheses only during collecting or even when it is finished!
Of course, there is a kind of research – descriptive, exploration – which is absolutely correct and still it does not relate variables to anything and thus it does not work with hypotheses as they were described above. Such research is used mainly when the theory about the given issue is poor and there is nothing to stick to, the problem is only being mapped and own theory is being created. In such a case, hypotheses are formed only after making the theory a system.
Forming hypotheses
To form hypotheses correctly, it is necessary to carefully follow three basic requirements (rules) for formulating hypotheses (violating of the rules is the most frequent cause of mistakes):
Hypotheses are statements and they must be formulated as declarative sentences. They must not be confused with research question (problem).
Hypotheses must express a relationship of at least two variables. Such relationship between two phenomena must be clearly and explicitly expressed. It is good to compare and verify variables: differences (more, more often, stronger, higher, different), relationships (positive, negative relationship, correlation) or consequences (the more – the less, if – then, both – and).
Hypothesis must have the ability to be tested. It must be possible to validate or invalidate a hypothesis. Variables must have the ability to be measured or classified (age: 16, 17…; sex: female, male; popularity: degree of popularity on a scale; quality of life – it is necessary to find life quality indicators).
Correct examples:
Differences: Schoolchildren in the second grade like school education more than students schoolchildren in the fifth grade.
Studying in a group supports the increase in students’ performance.
Relationships: There is a positive relationship between a child’s intelligence and the education of his/her parents.
Consequences: The more the teacher praises the students, the more the students learn.
The more cohesive the group is, the bigger is its influence on individual members.
Correct examples (the hypotheses expresses the relationship of two variables):
Girls perform better in language tests than boys.
(Two levels of one variable – sex)
Authoritative style of raising children develops creativity less than democratic style.
(Variables: style of raising children (democratic vs. Authoritative) and creativity – would be measured with a test)
The most frequent mistakes in forming hypotheses:
Formulation is too complicated and long. Complicated scientific formulations make the situation only worse: stick to the rule than less is sometimes more; hypothesis must be clear, easy and simple.
Hypotheses contain too many variables the relationship between which are not clear.
More hypotheses which do not belong under any main hypothesis; insufficient relationship between the hypotheses. It is often cause by insufficient overview in literature, ignorance of the findings so far or inexperience in dealing with hypotheses.
Hypothesis was not validated which was caused by improper selection of the hypothesis. The theory on which the hypothesis was based may not have been suitable, it should have been based on a different theory. However, sometimes a theory may be valid only under certain conditions (age, size of the city, quality of the teacher) and it is very much appreciated if this is specified.
Example of an incorrect hypothesis:
Music activity of elementary schoolchildren in the 2nd-7th grades increases when a set of music activities anchored in current conception of teaching is fully used when using the set, teachers activate students in a creative way, using methods, forms and means respective of their age which support effective development of their relationship to music and the personalities of the students generally.
Even segmenting the complicated sentence does not help: none of the statements contain two variables:
If the set of music activities is fully used according to the current conception of teaching, music activity of the students increases.
Teachers working in accordance with the current conception of teaching, activate students in a creative way.
When teaching music activities, the teachers used methods, forms and means respective of the students’ age.
These music activities support effective music development of a student.
These music activities support effective music development of students generally.
Example of correct forming of more hypotheses:
H: When teaching reading according to the book by Virgovičová, fewer mistakes appear in children’s reading than when teaching according to the book by Němčíková-Němčík.
H1: ………..fewer mistakes in reading letters………
H2: …………………………………............... syllables ……
H3: …………………………………............... sentences …..
Examples of incorrect formation:
Czech daily press pays little attention to issues connected with schools.
(Comparison must be done. “Little” cannot be measured, press – one, how can we define attention?? Number of articles, their significance, number of fundamental ideas in the text? – the length does not need to reflect this.)
Good teachers make use of humor in teaching.
It is a statement which must be transformed into hypotheses (express relationship): Teachers who make use of humor in teaching get better teaching results than teachers who do not.
There are more students from secondary “grammar” schools at universities than students from other types of secondary schools.
Clearly and explicitly define the relationship between two variables, find a suitable formulation! The meaning here is biased: it can be the number of students in a classroom; it can mean a context whether they are admitted to university; a number of students who have been admitted to university; it can mean that the number of students is not the same => the initial groups are not the same.
We suppose that suitable forms of cooperation between school and parents are not always used.
(2 variables – frequency, form of cooperation)
Rephrase: Suitable forms of cooperation between school and parents are less frequent than unsuitable ones. (Then it would be necessary to define which forms are suitable and which are not – form it as an operation.)