23 Methods of Child Study

K. Arockia Maraichelvi

epgp books

 

 

 

 

1. INTRODUCTION

 

The theories of child development in general guide us in predicting the behaviour/performance of a child, termed as hypothesis for a research. But sound research in child development needs many more things apart from theory and hypothesis. A researcher should decide on the nature of subjects, the sample size, what the subjects will be asked, when and how. It is only then relationships can be examined and conclusion can be inferred.

 

This module will provide you a better understanding of the most commonly employed research strategies for children in particular under three major heads namely

 

1.      Developmental designs

2.      Research designs

3.      Methods of data collection

 

2.  LEARNING OBJECTIVES

 

At the end of this module you will be able to

  • Comprehend the developmental designs adopted to study children.
  • Develop an insight into the various research designs used to study children and its shortcomings and
  • Understand the different methods of data collection

3.  DEVELOPMENTAL DESIGNS

 

To answer the question of how do children change in due course of time in a scientific manner, the researchers in the field of child development starts with a research question or hypothesis as any other research. Hypothesis is a research statement or question that would be tested by reflecting on what the researcher wants to find out. Consider the following example in understanding this part of the module.

 

“How do sibling relationships change among children in their 6th, 12th and 18th year”?

 

The first step in doing this research will be to determine the type of study that tells us the most about the changes that happen in their sibling relationships. While trying to identify the methods, a researcher comes out with two possible approaches as specified below

 

1.Do I want to observe the changes over an extended period within a single population of children in their 6th year and then in their 12th year and later in their 18th year?

2.  Do I want to compare the sibling relationship among different population of children in their 6th, 12th and 18th year at the same point of time?

 

The first approach is referred to as the Longitudinal method of studying children, and the second one is called the Cross – sectional method.

 

Hence the two approaches are defined as follows

 

Longitudinal study – It refers to a research design, wherein one single population is studied over and over again at different ages or periods of their life span.

 

Cross-sectional study – It refers to a research design, wherein various group of subject at different ages or periods of their lifespan are studied at the same point of time No let us know about the benefits and the purpose of each of these approaches.

 

3.1 Longitudinal study

 

It is an observational study, where the researchers are in continuous process of observing the same subject over a relatively longer duration. To return to the example specified above, the study would observe the needed subject/subjects in their 6th year, and then again when they attain their 12 year and consequently during their 18th year. So the study would take 12 years to get completed. Though it is a time consuming approach, there are certain benefits as given below

  • The possibility of detecting development or its related changes at the individual level over the span of research
  • As the subject is studied repeatedly at different age, the sequence of events could be easily traced
  • This approach is more likely to suggest cause and effect relationships, as it accountsfor the development of sibling relationship from 6th year to 18th year and changes that occur over time.
  • These type of studies could be conducted with only a small group of subjects, so generalization to larger population becomes difficult
  • The subject’s drop out rate is high and hence the data prediction becomes complex
  • By continuously studying a particular subject for a long time might influence the subject to react wisely to the testing situations and results in contamination of results.
  • Cohort effect is another major pitfall. Cohort effect occurs due to the cultural and historical changes on the consistency of findings. That is, while studying a group of children born at a particular period may not have the pattern of development when compared with children born in another period.

3.2 Cross- sectional study

 

In order to alleviate the pitfalls of longitudinal design, researches developed the cross– sectional design. As longitudinal approach, the cross-sectional design of studying children is also carried out by observing the subject. To make it simpler, the researcher records the information without manipulating the environment. The major success of this type of study is because of its scope of comparing different population groups at a single point of time. Hence the time spanned for this type of study is relatively short. Moreover the factors that influence changes over time (both historical and sociological) get minimized.

 

However the cause and effect relationship could not be traced by this approach, because the subjects sharing the same cultural and historical experiences but at different periods of lifespan would have caused the changes inspite of their age induced changes. Also this type of study that covers a wide range of age span, also faces the potential difficulty of cohort effect as in the longitudinal approach.

 

3.3 Longitudinal Sequential study

 

In order to overcome the potential difficulties of both the designs explained above, child development researchers tried to improve the developmental designs and arrived at a new design called Longitudinal Sequential design. This design is a combination of both longitudinal and cross- sectional approach.

 

Longitudinal Sequential design is a research design, wherein a group of subjects born at different times are followed over a short span of time. Hence the disadvantage of longer duration of longitudinal study could be conquered and moreover the cohort effect of both the designs is also minimized as careful comparative analysis of historical and cultural changes could be observed.

 

Turning to the same example of sibling relationships provided, the answer to both the questions could be answered by one design called longitudinal sequential design. Look at this figure.

Three children in their 6th , 12th and 18th were studied in the year 2006 with reference to their nature of sibling relationship and in the same way another three children during the year 2010.

 

The green lines indicate the comparison of children born at different periods, and thereby accounting to the longitudinal design

 

The blue lines shows the cross sectional comparison, say for example a child of six years is studied in two different periods, 2006 and 2010.

 

The pitfall of both the longitudinal and cross sectional design is the cohort effect that gets resolved in this new design. The red lines compares the 6 year old child studied in the year 2006 with a 12 year child studied in 2010 and 12 year old in 2006 with 18 year old in 2010.

 

4. RESEARCH DESIGNS

 

Research design refers to the overall plan of a research to test the hypothesis formulated. The type of design has to be decided by the researcher to test a set hypothesis with a possible greater certainty. Generally two major designs are adopted to study child behaviour. They are

 

1.      Correlational design and

2.      Experimental design

 

4.1 Correlational design

 

The design that involves the gathering of information from individuals without altering their environmental settings even at a minimal level and looks at relationship between variables is referred as correlation design.

 

Eg. Hypothesis – Do the mother’s interacting style influence the child’s intelligence?

 

In the correlational studies, the relationships are examined by a term called correlation coefficient, which describes of how the two variables are associated with one another. The correlation coefficient ranges from +1 to -1, where ‘0’ indicates no relationship between variables, and +1 and -1 refers to strong relationship. The sign +/- shows the direction of relationship. Positive sign shows positive correlation which means as one variable increases the other also increases. The negative sign explains negative correlation and indicates that as one variable increases the other decreases. For the above set hypothesis of two variables namely mother’s interacting style and child’s intelligence

  •  A Positive correlation would illustrate that both the variables move in the same direction. In other words, when mother’s interacting style increases, the child’s intelligence increases
  • A negative correlation becomes an evidence for the variables moving in opposite direction, which means that when the mother’s interacting style increases, the child’s intelligence decreases
  • Zero correlation confirms no effect between these two variables.

       Though  the  direction  of  relationship  between  variables  could  be  identified  by correlation studies, the cause and effect of this relationship cannot be ascertained. Because the evidence is not just enough to conclude whether it is only the mother’s interaction influence the intelligence level of the child, as the opposite is also certainly possible. An intelligent child’s behaviour would have prompted mother’s favourable interaction. Moreover the interaction effect would have been influenced by a third extraneous variable that had not been accounted in the study. In order to overcome the shortcomings of correlational study, experimental design of research came into practice.

 

4.2. Experimental design

 

Experimental design determines the cause and effect of a relationship. For an experimental study the researcher considers certain events and behaviour of the subject related to the study and classifies into two variables, namely independent and dependent variables.

 

Independent variable – The variable thought by the researcher to influence another variable Dependent variable – The variable expected by the researcher to be influenced by the independent variable.

 

By controlling and manipulating changes in the independent variable, the cause and effect of a relationship would be inferred. The changes in the independent variable can be done by subjecting the sample of interest to one or more conditions of treatment and the behaviour /performance could be compared based on the dependent variable.

 

Consider the example ‘Do the quality of interaction between adults influence a child’s emotional behaviour while playing with a known peer?

 

Here, the interaction between adults serves as the independent variable, wherein it could be manipulated by the researcher in two ways

 

1. In one setting two adults interact in a friendly way in a warm environment in a corner of a room

2.In another setting two adults get involved in n argument and expresses behaviour like shouting, slamming the door etc.

 

The child playing with a known friend becomes the dependent variable that undergoes no alteration.

 

The researcher observed that the children in a hostile setting showed aggressiveness and are found to be distressed when compared to the children in warm environment.

 

However the challenge on the part of the researcher in an experimental study is that he/she should take special care to control the unknown traits/characteristics of the subjects involved, as this would hamper the accuracy of the findings. In the study just explained, there is a possibility those children who were already exposed to unfriendly environment exhibits aggressiveness irrespective of the manipulated settings. Thereby, the study cannot conclude that it is only because of the independent variable the child’s emotional reaction changes.

   To overcome this problem two methods are generally used

 

1. Random Assignment – The subjects being picked up randomly increases the chance of equal distribution of the subject’s characteristics

2. Matching – The subjects are measured for behaviour of interest (Aggressiveness in the above said example) ahead of the actual research and then randomly assigned based on their level of aggression to every treatment condition.

 

5.  METHODS OF DATA COLLECTION

 

After looking into the benefits and pitfalls of the developmental designs in child stdy and after identifying a particular relevant approach to conduct the study, what is next?

 

The next step is to identify how to gather data or information from the subjects of interest. Some major methods of data collection are through

  • Observation
  • Self-Reports – Interview and Questionnaires
  • Case studies

  5.1 Observation

 

As the term indicates, the information is gathered by mere observation of the subject and recording his/her behaviour as it happens. This method is again divided into two, namely naturalistic and structured observation.

 

Naturalistic observation refers to observation of subjects in real world setting. In other words observation happens in natural setting.

 

For example observing a child of 6-8 years in his own class-room setting and recording the behaviour of the child with key reference to the number of times the child extends physical and emotional support to his friend in the form of sharing, helping, comforting, expressing kindness or sympathy serves as a good method of studying the child’s social development.

 

However, every child being observed doesn’t get the chance of expressing every form of social behaviour at a particular time. To bridge this gap of observation, one another technique called as structured observation got into practice.

 

Structured observation is the method, wherein the researcher observed the subject in a laboratory setting by controlling the environmental factors. In other words the researchers provides cue for the display of needed behaviour

 

For example, the researcher observes the helping nature of a child by providing a situation where in an adult spill a box of pencils and records the child’s response to the situation.

 

Though this method of observation provides researcher with more control over the natural setting, it has its own flaws, as we cannot expect a child to behave in the artificial setting as they do in the natural setting. Hence, observation could only provide us with certain information on how the children behave. It should be supported with certain other methods of child study to reason out the motive that lies behind their behaviour.

 

5.2 Self-Reports – Interview and Questionnaires

 

Self report method of child study is one method that could fill the gap of observation methods. These reports are methods that ask subjects to answer questions related to their capacities, beliefs, attitudes, feelings and experiences. The method varies from unstructured interview (eg. Clinical interviews) to relatively structured interviews (eg. Well formulated questionnaires and standardized tests).

 

Unstructured interview – One best example of unstructured interview otherwise called clinical interview was the Piaget’s method of questioning a 5 year old child about his understanding of dreams. He used a flexible, conversational style of questioning to prompt the child to express his ideas in detail and give a complete picture of his dream. Unstructured interview has got two advantages

 

1. It encourages the subject to express their thoughts in a way that has close proximity with their everyday thinking process

2. It provides more information in less time.

 

1. It should be checked for the accuracy of information provided by the subject as there is a possibility of making up of the answers for one or the other reason.

2. The subjects who are poor in their verbal ability cannot put forth their ideas clearly.

3.  The strength of flexibility of this method becomes a weakness as the questions would be framed differently for different subjects.

To eliminate these disadvantages the structured interview method is adopted.

 

Structured Interview – This method of collecting data has got a set of questions either formulated by the researcher or standardized tests and the subjects are questioned in the same order. This technique is considered to be efficient to the clinical interview method as information from many subjects can be procured at the same time. Also, the answers have a chance of being tabulated by computers and subjected to consolidation and analysis easily. However problems of depth in information and accuracy of data arises.

 

5.3 Case studies

 

It is a method of collecting data, wherein a particular subject of interest is studied comprehensively to obtain a great deal of information. Case studies are often in clinical researches as their approach cannot be generalised to larger population. This method gives descriptive case details that offer important insights into the understanding of the subject’s behaviour.

 

Though case study is an elaborate method, the conclusions drawn from studying a subject cannot be applied to another subject as every individual varies. Also the chance of the investigator being biased in the interpretation of data is high.

 

6. ETHICAL ISSUES IN RESEARCHES INVOLVING CHILDREN

 

As the expedition for scientific knowledge widens day by day, the possibility of exploiting people is on the rise. It is here where the ethical issues and concerns gained significance. As we are concerned with the study of children the ethical concerns seems to be more complex than any other research areas. Certain concerns that need to be addressed are as follows

 

1. Children are more vulnerable to physical, mental and psychological harm in comparison to adults.

2.Immature mentality of the children poses a big challenge of being assessed or evaluated.

 

American Psychological Association and Society for Research in Child Development had come up with special ethical guidelines for research with children to address the above said issues. Some of the basic research rights drawn from these guidelines are important to be dealt in this module. They are

  1. Protection from harm – when any sort of physical, mental or psychological harm seems possible, an alternate method of procuring data should be considered
  2. Informed Consent – Consent of significant others of the child subject (Such as parents or school officials) should be obtained in writing. Moreover in case of unwillingness after the consent, the child has the right to be freed from the research at any point of time of the research
  3. Privacy – the identity of the subject shall be concealed if the subject feels so.
  4. Knowledge of results – the child subject possess the right to be informed of the research result and its interpretation
  5. Beneficial treatment – children of control group as well have the right to receive the experimental treatment if found to be beneficial.

Hence, researchers first and foremost responsibility is to foresee the potential risks that the children would encounter as an effect of the study conducted on them. Then follows the efforts to minimize those risks. The major responsibility is to make sure that the benefits of research outweigh the potential hazards.

  1. CONCLUSION

Thus this section had given you an elaborate insight of the various methods of studying children and its relevant research designs. The advantages and pitfalls of each method and design were discussed over the other, which would prompt you to consider the best method of studying children.

you can view video on Methods of Child Study

Web links

  • www.open.edu/…/childhood…/childhood…studies/childhood/methods-studying-children…
  • www.psy.cmu.edu/~rakison/POCDclass2_2006.htm
  • psych.colorado.edu/~colunga/p4684/methods.pdf
  • www.psychologydiscussion.net/child-psychology/studying-child…4…methods/943