Are you still unsure what thesis topic to pursue in the course of honing your skills in research? Or maybe you have already one but you are in doubt if it is a good research problem at all? Read on to find out if you’re on the right track.
One of the difficulties that graduating undergraduate and graduate students encounter in the course of preparing their thesis involve the proper selection of their research problem. If you are one of them, how should you go about it? How will you know that the research problem is good enough for you to spend your time, money, and effort?
A Good Research Problem
You will find out that a good research problem is a good one if you have critically assessed and satisfied at least two things:
Are there available methods to carry it out?
Do you have the resources to pursue it?
Take a look and see if you have these things at your disposal. If not, then, you are not yet ready to do anything at all.
In doing any kind of research, you need to use a standardized method so that other researchers can confirm your findings or you are able to compare your results with someone who pursued the same topic as yours.What methods have been used so far to allow the resolution of the research problem?
This requires you to browse the literature and look for appropriate methods on the topic at hand. Reviewing the literature will also help you keep up with current topics, identify which ones need more information or lack data for better understanding of a given phenomenon.
Reviewing the literature doesn’t mean reading one or two scientific papers but several, or a lot, so that you are able to synthesize or point out what has been missed, inconsistent findings, or different conclusions between authors. This just means that studies on that topic has not yet matured and that further studies need to be conducted to resolve pertinent issues.
Resources refer to things such as available time to complete the research, the amount of money you will need to do the activities associated with your research, your capacity as a researcher, availability of equipment needed for data collection and analysis, among others.
Attending conferences and talking with experts on a current issue can also guide you in your thinking. Is there local interest to pursue the research issue? If there is, then you can take off from where others have left and discover new relationships between variables.
Once you find a good research problem where you are able to state it clearly and well, you can consider half of your work done. You are well on your way towards becoming an accomplished researcher.
Is there a way to simplify the preparation of your research or thesis proposal without leaving out the important items to include in its preparation? Try the matrix approach described here and reap the benefits.
You may find yourself getting into the trouble of writing and rewriting your thesis proposal because you tend to miss important details pertinent to what you intend to investigate on and how you are going to go about it. Research or thesis proposal preparation is very time-consuming and can cause undue worry especially if you have set a fixed time frame to finish your thesis. If your desire is to have your research proposal approved soonest so you can start gathering the data you need, this is for you.
A systematic way of ensuring that everything is well addressed or covered fully in your research paper is possible with the use of a matrix. This technique is most appropriate when you want to make sure that you have adequate preparation, especially the appropriate methods to use, to answer the research questions.
What is a matrix?
My students would mull at me every time I tell them about using a matrix to do their research work in a more ordered, straightforward or effective manner. This is a not-so-common technical term to most of them. Although they usually wouldn’t ask, I follow-up with an explanation of what a matrix means.
I would then scrounge for a clean sheet of paper or anything that can serve the purpose to illustrate how a matrix can be used to set one’s mind into focus. A matrix is basically a table with rows and columns. The technique works this way:
1. Prepare a table with the following headings for each column:
You may fold the sheet of paper into three equal-sized columns or draw a line downwards to separate each column.
2. List the research questions
Under the heading “Research Question,” write the series of research questions that you intend to pursue in their logical order. Logical order means that you arrange research questions chronologically. It is ordered in such a way that answering the first question will facilitate the resolution of the next question.
3. Supply the required methods to answer the research questions
Under the heading “Methodology,” look at the left column and think how you would go about answering the research question. What shall you use to provide the information required in the first question, the second one, third, and so on. As you finish writing the method to use, place a line beneath to separate the questions and their corresponding methods from each other.
4. Select the appropriate statistical tool
Under the third column with the heading “Statistical Analysis,” recall your statistics lessons or consult a statistician about the correct statistical tool to analyze the facts to be gathered in the study. Does the research question need simple descriptive statistics such as mean, median, mode or percentages? Or do you need to apply a correlation analysis, a test of difference between means, or a multivariate analysis? You can also add under this column the corresponding graphs or tables that you will need for better discussion of the findings.
Now, guided by your matrix, you will be able to answer your research questions with confidence. You make sure that everything is covered by setting a one-to-one correspondence in the crucial elements of the research proposal, i.e., the research questions, the methodology, and the corresponding statistical analysis.
An example is given below to show it should look:
That wraps it up. Try it and be more systematic in preparing your research proposal.
When not designed properly, data obtained using the survey method is next to useless. Find out why.
While a lot of information can be gained from surveys as it is easy to get large samples using the method, results should be taken with caution because of its inherent weaknesses. This is the reason data analysts do not treat data obtained from surveys in the same way as those obtained using other means, that is, those that do not rely on people’s opinions or subjective judgment.
So what are the weaknesses of the survey method? Below is a list with brief explanations.
Weaknesses of the Survey Method
1. Respondents protect their interest
There is really no guarantee about the truthfulness of the respondent’s answers. When dealing with sensitive or controversial issues, there is a tendency among interviewees to avoid answers that may be detrimental to their interests. They may not even answer the question at all.
2. Attitude is different from behavior
As the famous expression goes: “Do what I say, not what I do.” A considerable number of research on the relationship between attitude and behavior has demonstrated that there is no correlation between what people say they would do than their actual behavior.
In a recent study in the United Kingdom, for example, 99% of people interviewed said they had washed their hands after using the toilet. The truth is, as revealed by electronic recording devices, only 32% of the men and 64% of the women actually did it.
People know what is expected of them, pretend that they do it when actually they don’t.
3. Responses are taken lightly
Some respondents are not really predisposed to answering questions especially if it involves answering a series of long questions. Similarly, the interviewer may not be very keen about the data collection.
4. Poor memory among respondents
Interviewees tend to forget past events, places, or experiences. Their answers, therefore, are mainly estimates or rough approximations as their memory would allow them to recall those things.
Would you rely on information recalled 20 years ago by a 60-year-old respondent? Of course not.
5. Answers can be manipulated
Some interviewees answer the questionnaires to gain a favor. People answer what they believe the interviewer wants to hear. This threatens the validity of the answer.
What Then Should Be Done?
To get the most from using the survey method, particular attention must be given to the design of the questionnaires,selection of samples, and administration. For greater reliability and validity, other methodologies are applied to confirm the findings of a survey (see Triangulation).
1. Winterman, D. (2012, October 15). Handwashing: Why are the British so bad at washing their hands? Retrieved September 2, 2013, from http://www.bbc.co.uk/news/magazine-19834975
How do you come up with your research design? Here are two examples of blood pressure exploratory studies as leads toward research design development.
Blood Pressure Exploratory Study
I find the practical aspects of applying research enjoyable and designed experiments to uncover some relationships or to resolve my problem.
Several years ago, I convinced my doctor to cut my blood pressure drug maintenance. I simply presented to him a graph I prepared using a spreadsheet application and an Analysis of Variance (ANOVA) to compare my blood pressure readings with the full dosage of the drug, half of it, and a fourth of it. I also compared the two groups at a time using t-test and I got the same results. The graph and the statistical analysis showed my blood pressure readings did not show a significant difference as I gradually reduced the dosage of the prescribed drug.
The Diet Experiment
The primary purpose of the above experiment is to see whether diet can produce the same results as a drug in lowering blood pressure. One of the active components of the drug in question was potassium. I thought it is better to take natural food to get the mineral. I computed the amount of potassium that corresponds to the dosage by eating a number of potatoes that will supply such amount and gradually reduced the drug dosage while monitoring my blood pressure daily.
From my musings on the nutritional value of potato, it would take about two to three medium-sized potatoes to get the same amount of potassium. So, I took at least three potatoes a day to correspond to the amount of potassium required as I reduced on the drug dosage.
When I came around 1/4 of the prescribed dose, I asked the doctor if I can forego the drug because I suspect it is one of the reasons why I was feeling weak. And I confirmed it by finding information on the side effects of the drug. The doctor was amazed because I reduced the drug to such a small amount that, according to him does not anymore provide substantial benefit in lowering blood pressure. My blood pressure has stabilized. He said I can forgo the drug and return to him when my blood pressure rises again.
Despite this apparent success in my desire to live without the drug in my system, I do not recommend this approach to anybody because it might work differently to different people. I take certain precautions when I do conduct studies on myself.
Blood Pressure and Exercise Experiment
Recently, I’m at it again. This time, I just would like to verify if indeed exercise provides the benefit of lowering blood pressure. My readings say so and I would like to personally find out what the numbers will show. I monitored by blood pressure before exercise, right after exercise and 15 minutes or more after my exercise so that my blood pressure will stabilize at rest.
I just started this last week and saw a trend just from three readings. I show the results of the blood pressure monitoring in the table below.
The results are interesting because obviously, my blood pressure went down right after exercise and dropped more 15 minutes after. Upon waking up, my blood pressure readings show that my systolic reading indicating the maximum arterial pressure is higher than the normal 120mm Hg but the systolic reading is normal. After exercise, the systolic pressure reduced greatly just by visual inspection even while the heartbeat is high. After 15 minutes, the systolic and diastolic readings even went down further while my heartbeat approximates its normal value.
So are the results conclusive enough that exercise lowers blood pressure? There is no doubt exercise lowers blood pressure but I have not seen details on how much blood pressure is reduced by exercise. This data informs me right away the benefits of exercise and serve as an encouragement to engage and maintain my exercise routine.
Today, when I went for my usual six kilometer run in 41 minutes and 38 seconds (the fastest so far in that distance), my blood pressure after exercise approximates the previous values. It is 104/65 with a heartbeat of 95. Again, after my heartbeat stabilized 15 minutes later at 64, while my corresponding blood pressure is 101/59.
Data Collection Procedure for the Exercise Experiment
How did I come up with such values? What is the data collection procedure? I collected this data systematically, making consistent readings as much as I can. Roughly, the data collection procedure goes this way and can be replicated by anybody.
Record BP and heartbeat –> stretching exercise for five minutes –> slow walk of 8 minutes –> run proper –> cooling down with a slow walk for about 15 minutes –> five-minute stretching –> record BP and heartbeat –> rest for 15 minutes –> record BP and heartbeat
I used an Omron Automatic Wrist Blood Pressure Monitor in making blood pressure readings. It read a little higher than the standard sphygmomanometer but it reads consistently. So it will be easy to calibrate it for more standardized results.
From this exploratory study which confirms the benefits of exercise in a quantitative way, a research design may be developed for more rigorous analysis. You should notice, however, that sleep may also have an influence on blood pressure readings so I marked the third reading with an asterisk. The quality of my sleep in the first two readings is not that good as I only had six or less hours of sleep while on the third reading, I got quality sleep of seven hours or more. This apparently resulted to lower blood pressure readings upon waking up.
This means that if I pursue this experiment, I should make my measurements consistent and consider the hours of sleep and factor it in for analysis. I should also make sure that monitoring time should be the same all throughout the duration of the study.
Now, the question is: “Are there studies conducted like this before?” I actually don’t know as in truth I am not a medical researcher. At best, my experience is only a case study; a description of my case. But a review of literature will tell me if a similar experiment was done by anyone on a greater number of people. These experiments are guided by different theories on the effects of exercise to health developed through time. In my case, I just did it out of mere curiosity to verify my readings which are also backed by theories.
From these initial data, an experimental research design may be developed to ensure that the evidence obtained answers the questions initially posed for the study. Two questions were posed in these two examples: 1) Can a well-planned diet produce the same results as a drug in lowering blood pressure?, and 2) Does exercise lower blood pressure?
From simple case studies like these, experiments may be designed to test if the findings are consistent for a greater number of people. This will also provide insights on which variables should be included for analysis.
1. Mayo Clinic (n.d.). Exercise: A drug-free approach to lowering high blood pressure. Retrieved August 27, 2013, from http://www.mayoclinic.com/health/high-blood-pressure/HI00024
Have you heard or read about a new mammal species discovered in Mount Andes in South America? It is called olinguito. The bearcat resembles it. See the difference.
As I browsed the internet for current information on science, I bumped on the Science Daily website featuring olinguito (Bassaricyon neblina), a mammal species of mistaken identity. What first struck my attention is that olinguito resembles the mammal I took a shot two weeks ago in a wildlife conservation facility. It is called the Palawan bearcat.
Olinguito vs Bearcat
Examining the picture of the newly discovered mammal species closely, I could not help but compare it with the bearcat (Arctictis binturong whitei) which is found in Palawan. Accordingly, the olinguito looks like a cross between a house cat and a teddy bear.
The Palawan bearcat is similarly described. It is also a cross between a bear and a cat but it neither belongs to the bear nor the cat family. The mammal belongs to the family Viverridae.
For copyright reasons, I could not post a picture of the olinguito for comparison here, but this is available in the ABC News website. Indeed, the olinguito looks very much similar to the mammal found in the western part of the Philippines, in the island of Palawan.
You may click the link to the ABC News website to see a picture of olinguito that bears close resemblance to the bearcat I show below.
Feeding Habit of the Bearcat vs the Olinguito
The bearcat is sometimes treated as a distinct species although it looks very much like the ones found in Borneo. It has a prehensile tail, meaning, its tail can grasp an object. This is because the bearcat lives in tree canopies to feed on both plant material and other smaller animals like insects, rodents, birds and even fish. Thus, it is omnivorous (animal and plant eater) as opposed to olinguito which is a carnivore (exclusively flesh-eater).
The information on olingo’s feeding habit, however, is a bit confusing because the headline in Science Daily says it is a carnivore. But scrolling down further in the website, it reports that during the three-week expedition of Helgen and Kays, they noted that it is a nocturnal animal and feeds mainly on fruits.
This requires further study as the discovery of this new species spurred interest on finding out more about its characteristics, habitat, distribution and conservation. More expeditions are being planned at the moment.
Behavior and Size Comparison
I find the bearcat much cuter than the olinguito. In fact, tourists love to have the tamed bearcat on their shoulders for a quick shot every time they visit the Palawan Wildlife Refuge and Conservation Center (PWRCC) where several of them are kept in captivity. Placing the bearcat on one’s shoulders, however, is a dangerous practice because the bearcat can rip flesh easily just like a bear.
More than a decade ago, I saw someone made the mistake of handling it when it was already big and got slashed by its razor-sharp claws in the process. It is docile when tamed but accidents do happen.
The bearcat grows up to 1.4 meters and weigh more than 20 kilograms as opposed to olinguito’s two pounds. This probably is the reason why olinguito was named so. It is a small olingo.
A Demonstration of Keenness
New findings like this arise from a keen sense of observation which good researchers must possess. For many years, the olinguito was regarded as a species of kinkajou, another mammal living within the same geographic range. But it took a team of scientists headed by Helgen, who noticed olinguito’s smaller teeth and skull in the museum collections, to uncover a feature that everyone overlooked. Keenness is the key.
1. Wilson, D. E.; Reeder, D. M. Mammal Species of the World (3rd ed.). Johns Hopkins University Press. pp. 548–559. Retrieved August 16, 2013, from http://www.departments.bucknell.edu/biology/resources/msw3/browse.asp?id=14000270
2. Widmann, P., De Leon, J. & Duckworth, J.W. 2008. Arctictis binturong. In: IUCN 2013. IUCN Red List of Threatened Species. Version 2013.1. . Downloaded on 16 August 2013.
3. Smithsonian Institute (2013, August 15). New species of carnivore looks like a cross between a house cat and a teddy bear. ScienceDaily. Retrieved August 16, 2013, from http://www.sciencedaily.com /releases/2013/08/130815143101.htm
How do you extract useful information from a group of people in connection with your research? One of the tools used is focus group discussion. Read on to find out how this is done.
If you engage in social research or study research methodologies, one of the common (sometimes abused) methods of data collection that you should be familiar with is focus group discussion or FGD. Aside from soliciting ideas that will help answer or narrow down your research topic, the output of the discussion verifies or confirms the results of surveys designed to answer research questions that you are interested in (see Triangulation).
What is focus group discussion, when do you use it and how should you conduct it? What good practices should be observed? This article provides answers to these questions.
Definition of Focus Group Discussion
Sometimes, FGD is also called focused group discussion because the discussion focuses on questions that seek multi-stakeholder response. It may also refer to the ‘focus group,’ that is, those who are found relevant to take part in tackling the issues raised by the researcher.
Essentially, FGD is a discussion of issues and concerns between a selected group of four to eight people. It serves as a venue to confirm and verify the participants’ viewpoints and draw out their experiences so that they are able to build a consensus about the research topic. A well-trained moderator guides the progress of the discussion using a set of questions prepared by the researcher.
How are the Participants of FGD Selected?
The participants of the FGD are selected using a set of guidelines or criterion such that the participants are able to give useful or relevant information to meet the objectives of the study. This requires familiarity with the background of the participants. It is, therefore, common practice that managers, leaders of the community or those who have lived in the community for a while or someone familiar with the business of an organization are consulted before conducting an FGD.
For example, in a study on coastal resource use, if the issue relates to dynamite fishing in the coral reefs, the participants to look for include representatives from the groups of fishers, fish traders, former dynamite fishers, law enforcers, explosives suppliers, local policy makers, non-government organizations or associations, among others who have direct or indirect transactions in the community. Avoid bias in the selection of participants such as including only those who are accessible or favoring a certain political group.
Example FGD Questions
Examples of questions that relate to illegal fishing in the coral reefs that will serve as the focus of the FGD are the following:
What are the target fish species of the illegal fishers?
How do the illegal fishers get their explosives?
How much do the dynamite fishers earn from their activity?
Why do the dynamite fishers engage in this illegal activity despite prohibition?
What are the risks associated with dynamite fishing?
At what time of the day and how frequent do the dynamite fishers go out to fish?
Where are the dynamite fishers coming from?
Of course, the questions will ultimately depend on what information you would like to draw out from the participants. The FGD enables you to explore which variables you will include and focus the quantitative (if ever) part of your study.
You might want to relate fisher income with frequency of dynamite fishing. Or you might want to quantify the costs and benefits of dynamite fishing (taking the point of view of the fisher). The end justifies the means, so they say.
How to Conduct the FGD
The following are needed resources to conduct a focus group discussion.
A trained moderator or facilitator. The moderator may not necessarily be the researcher himself but someone familiar with the issues to be discussed. Hence, he should confer with researcher before conducting the FGD process. He should have a good background knowledge of the participants and must not involve himself in the discussion, such as arguing with the participants. His main role is to introduce and explain the questions, clarify issues raised, confirm responses, encourage expression of ideas, among other related functions. He summarizes the process at the end of the discussion.
A note taker. The note taker records the progress of the FGD. He does not only list the oral expression of ideas of the participants but also their actuations or non-verbal expressions. He clarifies points once in while by getting the moderator’s attention on points that are not clear. He furnishes a copy of the transcripts to the participants as a matter of transparency.
The quality of information gathered through the FGD depends to a large extent on the skill and keenness of the moderator and the note taker. For best results, rapport between the researcher’s group and the participants should be made such that the participants will not inhibit themselves from freely expressing their ideas.
The following materials should be made available during the conduct of the FGD:
Recording material. The standard note pad and pencil or pen must always be available. Although laptops, tablets, cameras, MPEG recorder, or anything electronic will work in an urban environment, a different situation exists in FGDs conducted in far-flung areas. Although these gadgets may be used to record data in the field, these are prone to many problems such as low batteries, broken during the trip, got submerged and damaged while wading a river, among others. If electronic data recording equipment is really desired, then these should be weather and/or shock resistant.
Group memory. Group memory is something that the participants can refer to as the discussion takes place. The participants focus their attention towards this attention-getting list of questions and responses. This could be a set of Manila paper with pre-written questions, a whiteboard or blackboard, or a mini-projector if you may.
Attendance sheet. If you do research for somebody (say as a consultant) or in compliance with your thesis requirement, you need this because it serves as evidence that you really did the FGD. This will also help you find your respondents if you will need to go back and clarify points.
Global Positioning System (GPS). This will aid you in locating the place where you did the FGD. This is good information to those who would like to make a follow-up study in a similar place.
The information derived from the FGD, aside from fulfilling an academic requirement, is useful in policy making and management. It can lead to agreement on certain controversial issues and evaluation of program or project accomplishments in the target community.
International Institute of Rural Reconstruction (1998). Participatory methods in community-based coastal resource management.
In conducting research, being partial can lead to faulty conclusions. This tendency is conveniently called bias.
How can a researcher avoid committing this blunder? This article explains what bias is and suggests ways on how to reduce it.
One of the important considerations in research involving people’s response (i.e., social research) is to reduce or eliminate researcher bias. If a researcher conducts the investigation in a biased manner, research outcome becomes inaccurate and unreliable.
If the results of a study are unreliable, inaccurate or invalid, questions will arise on many fronts and doubts regarding the conclusions of the study will compromise its publication in a refereed journal. In businesses, decision-making based on the results and conclusions of the study will be faulty thus lead to profit loss or inability to solve organizational or operational problems. Bias can distort the truth.
But what causes bias and how can you as a researcher avoid it?
Definition of Bias
A relevant definition of bias in the Bing dictionary states thus: “bias is an unfair preference for or dislike of something.” In the context of research, this means that the researcher does something that favors or skews towards a certain direction. The researcher may deliberately or inadvertently commit it.
Causes of Bias and Suggested Solutions
While it will be difficult to eliminate all sources of bias in the conduct of your research totally, being aware of the following common pitfalls in the practice of research is desirable:
1. Personal convenience in data collection
Many of those who conduct research fail to do good research because they want to do it at their convenience. For example, instead of getting a random sample of respondents, he may just interview anyone that gets in his way. This approach is not an objective way of getting a good sample from a given population of the study.
Ask yourself the question: “Am I doing this part of the research for my personal convenience?” If you are, then recognize that this will introduce bias and reduce research quality.
Select respondents randomly.
2. Favoring your own stand
While the nature of your research may be argumentative, favoring a preconceived position on the subject you are investigating will cause bias in your results. You will have the tendency to steer the results of your study to the direction that you want.
Stick to what your data shows.
Do not manipulate the results.
3. Inadequately prepared questionnaires
Inadequately prepared questionnaires can lead to many biases in the results of your study. Make sure that you follow good practices in preparing questionnaires.
Read reliable references on how to prepare questionnaires objectively.
Once you are ready with your questionnaire, validate it to a non-respondent group.
Make changes when appropriate.
This situation means that you should take enough time to prepare your questionnaire. It is better to start right than having doubts about the reliability, validity or accuracy of your data later on.
4. Faulty Data Collection Procedure
Watch out for unreliable answers to the questions that you posed. It is possible that the following occurred during the data collection:
The whole point of this discussion is that the researcher should take all precautions against doing things that may impact negatively or threaten data accuracy and reliability. A researcher, therefore, must be neutral and objective-minded in carrying out his study.
This article describes market research, its goals and objectives, the kind of data gathered and how those data are analyzed. An example is provided.
As my students in research come from various disciplines, I need to strike a balance on the topics and generalize as much as possible on the principles of research. I take effort, however, to have specific examples relevant to my student’s background, so that they will appreciate better the role of research in their respective fields.
Some of my students are business graduates and many of them cannot imagine how research works in their field. The truth is, research is very much relevant in the field of business, particularly the marketing aspects that this article focuses on.
I see it necessary, therefore to define market research, what are is its goals and objectives, what kind of data is gathered and how those data are analyzed.
Definition of Market Research
After reading through the many definitions of market research, I came up with the following definition relevant to this discussion. Market research is simply the process of using research in view of increasing sales or income within the shortest time possible to gain the greatest profit.
Goal and Objectives of Marketing Research
The main goal of any business is to achieve greater or more sales, greater productivity, greater or faster return on investment (ROI). Marketing research could provide the information required to realize this goal.
The objectives of marketing research most commonly revolve around the following interests:
find out which products are preferred by consumers
determine which group of people buy which type of product
where the buying customers live
what age group search for and buy which product
how long people stick with the product they buy
at what time or periods do people buy which product
what do people like and dislike
how much are people willing to pay for a good or service
a lot more…
To simplify matters, marketing research, essentially just want to find out the characteristics of consumers and see how they should design products, improve services, develop strategies or techniques to capture these customers. A business thrives if it is able to answer the needs and wants of its customers. This makes it competitive.
If a company does not understand its customers, then most likely it will suffer a great loss in sales or reduce their income, spend more than they earn, and eventually get bankrupt. They may be promoting services or selling and producing products which are not relevant to customers’ needs and wants. Why produce a product that does not sell anyway? Why offer a service that is not in demand? And why keep on operating if the business is losing?
How the Results of Marketing Research are Analyzed
The data gathered about customers is useless unless analyzed using advanced statistical applications. Various approaches are applied in analyzing data obtained from market research but the common approaches employ multivariate statistics such as multiple regression, factor analysis, canonical correlation, multiple discriminant analysis, among others.
What is multivariate statistics and why multivariate? Multivariate statistics simply refers to analysis using not just a few variables; not two, or three but several or many variables at the same time. Aside from saving time, the results of such analysis can pinpoint which customer characteristics really matter when it comes to product purchase or sales.
To further clarify this idea, a market researcher might ask: “Which of the following customer characteristics: age, gender, occupation, income, residence, or nationality click more often the ads on an electronic product?” This information may be sourced from data gathered when someone signs up for a service such as an email or a social networking site. When you sign up in whatever services that are given free in the internet, bear this in mind. There is no such thing as free lunch.
Those who click on the ads on electronics are potential customers and knowing their characteristics will help sellers focus their marketing strategies. A multiple regression analysis will show this information. But of course, marketing researchers would have to find out if indeed there is a correlation between clicks and sales. A simple correlation analysis will enable them to answer this question.
If the data analysis reveals that gender has something to do with interest on an electronic product, then the product sellers should design marketing strategies that consider the role that gender plays on product selection. What products do men and women want? On the other hand, age may also be a factor, so this must also be considered. A model can then be constructed to estimate demand for the product using a combination of factors that predict its sales.
Well, this is just an introduction to market research with a simplified example. The point is that market research is a crucial component of business strategy especially among large businesses. A little change in the practices of a large company or even a small business can mean a lot in sales. The information provided by market research is an important part of business decision making.
In the research and statistics context, what does the term model mean? This article defines what is a model, poses guide questions on how to create one and provides simple examples to clarify points arising from those questions.
One of the interesting things that I particularly like in statistics is the prospect of being able to predict an outcome (referred to as the independent variable) from a set of factors (referred to as the independent variables). A multiple regression equation or a model derived from a set of interrelated variables achieves this end.
The usefulness of a model is determined by how well it is able to predict the behavior of dependent variables from a set of independent variables. To clarify the concept, I will describe here an example of a research activity that aimed to develop a multiple regression model from both secondary and primary data sources.
What is a Model?
Before anything else, it is always good practice to define what we mean here by a model. A model, in the context of research as well as statistics, is a representation of reality using variables that somehow relate with each other. I italicize the word “somehow” here being reminded of the possibility of correlation between variables when in fact there is no logical connection between them.
A classic example given to illustrate nonsensical correlation is the high correlation between length of hair and height. It was found out in a study that if a person has short hair, that person tends to be tall and vice-versa.
Actually, the conclusion of that study is spurious because there is no real correlation between length of hair and height. It so happened that men usually have short hair while women have long hair. Men, in general, are taller than women. The true variable behind that really determines height is the sex or gender of the individual, not length of hair.
At best, the model is only an approximation of the likely outcome of things because there will always be errors involved in the course of building it. This is the reason why scientists adopt a five percent error standard in making conclusions from statistical computations. There is no such thing as absolute certainty in predicting the probability of a phenomenon.
Things Needed to Construct A Model
In developing a multiple regression model which will be fully described here, you will need to have a clear idea of the following:
What is your intention or reason in constructing the model?
What is the time frame and unit of your analysis?
What has been done so far in line with the model that you intend to construct?
What variables would you like to include in your model?
How would you ensure that your model has predictive value?
These questions will guide you towards developing a model that will help you achieve your goal. I explain in detail the expected answers to the above questions. Examples are provided to further clarify the points.
Purpose in Constructing the Model
Why would you like to have a model in the first place? What would you like to get from it? The objectives of your research, therefore, should be clear enough so that you can derive full benefit from it.
In this particular case where I sought to develop a model, the main purpose is to be able to determine the predictors of the number of published papers produced by the faculty in the university. The major question, therefore, is:
“What are the crucial factors that will motivate the faculty members to engage in research and publish research papers?”
Once a research director of the university, I figured out that the best way to increase the number of research publications is to zero in on those variables that really matter. There are so many variables that will influence the turnout of publications, but which ones do really matter? A certain number of research publications is required each year, so what should the interventions be to reach those targets?
Time Frame and Unit of Analysis
You should have a specific time frame on which you should base your analysis from. There are many considerations in selecting the time frame of the analysis but of foremost importance is the availability of data. For established universities with consistent data collection fields, this poses no problem. But for struggling universities without an established database, it will be much more challenging.
Why do I say consistent data collection fields? If you want to see trends, then the same data must be collected in a series through time. What do I mean by this?
In the particular case I mentioned, i. e., number of publications, one of the suspected predictors is the amount of time spent by the faculty in administrative work. In a 40-hour work week, how much time do they spend in designated posts such as unit head, department head, or dean? This variable which is a unit of analysis, therefore, should be consistently monitored every semester, for many years for possible correlation with the number of publications.
How many years should these data be collected? From what I collect, peer-reviewed publications can be produced normally from two to three years. Hence, the study must cover at least three years of data to be able to log the number of publications produced. That is, if no systematic data collection was made to supply data needed by the study.
If data was systematically collected, you can backtrack and get data for as long as you want. It is even possible to compare publication performance before and after a research policy was implemented in the university.
Review of Literature
You might be guilty of “reinventing the wheel” if you did not take time to review published literature on your specific research concern. Reinventing the wheel means you duplicate the work of others. It is possible that other researchers have already satisfactorily studied the area you are trying to clarify issues on. For this reason, an exhaustive review of literature will enhance the quality and predictive value of your model.
For the model I attempted to make on the number of publications made by the faculty, I bumped on a summary of the predictors made by Bland et al. based on a considerable number of published papers. Below is the model they prepared to sum up the findings.
Bland and colleagues found that three major areas determine research productivity namely, 1) the individual’s characteristics, 2) institutional characteristics, and 3) leadership characteristics. This just means that you cannot just threaten the faculty with the so-called publish and perish policy if the required institutional resources are absent and/or leadership quality is poor.
Select the Variables for Study
The model given by Bland and colleagues in the figure above is still too general to allow statistical analysis to take place. For example, in individual characteristics, how can socialization as a variable be measured? How about motivation?
This requires you to further delve on literature on how to properly measure socialization and motivation, among other variables you are interested in. The dependent variable I chose to reflect productivity in a recent study I conducted with students is the number of total publications, whether these are peer-reviewed or not.
Ensuring the Predictive Value of the Model
The predictive value of a model depends on the degree of influence of a set of predictor variables on the dependent variable. How do you determine the degree of influence of these variables?
In Bland’s model, all the variables associated with those concepts identified may be included in analyzing data. But of course, this will be costly and time consuming as there are a lot of variables to consider. Besides, the greater the number of variables you included in your analysis, the more samples you will need to obtain a good correlation between the predictor variables and the dependent variable.
Stevens recommends a nominal number of 15 cases for one predictor variable. This means that if you want to study 10 variables, you will need at least 150 cases to make your multiple regression model valid in some sense. But of course, the more samples you have, the greater the certainty in predicting outcomes.
Once you have decided on the number of variables you intend to incorporate in your multiple regression model, you will then be able to input your data on a spreadsheet or a statistical software such as SPSS, Statistica, or related software applications. The software application will automatically produce the results for you.
The next concern is how to interpret the results of a model such as the results of a multiple regression analysisl. I will consider this topic in my upcoming posts.
A model is only as good as the data used to create it. You must therefore make sure that your data is accurate and reliable for better predictive outcomes.
Bland, C.J., Center, B.A., Finstad, D.A., Risbey, K.R., and J. G. Staples. (2005). A Theoretical, Practical, Predictive Model of Faculty and Department Research Productivity. Academic Medicine, Vol. 80, No. 3, 225-237.
Stevens, J. 2002. Applied multivariate statistics for the social sciences, 3rd ed. New Jersey: Lawrence Erlbaum Publishers. p. 72.
One of the common concerns of graduate students is finding topics relevant to their field of specialization. If you are a graduate student of education, here is a list of recent issues and concerns on education that you might want to focus on as your research topic.
Here are four examples of research topics on education. The list focuses on children as vulnerable members of society.
1. How children react to loss of parental care
There are many instances that children have to cope with loss of parental care as a result of death of their mothers, separation of parents, assignment of a parent to distant places, among others. This feeling of loss is generalized as grief. Although it is easy for adults to adjust to such circumstances, questions arise as to whether children are able to cope.
Example Research Questions:
What are the causes of grief among children?
How do babies express their grief?
How does grief affect children in various developmental stages?
How do children cope up with the loss of a loved one?
Is there a difference in the behavior of children who lost their mothers compared to those who did not?
2. Early childhood development
Many mothers, especially those who are new, are very much concerned about the development of their children. Nowadays, there are so many stresses in the environment that could threaten the development of children. The following are questions on child development that research could help shed light on.
Example Research Questions
What are the early warning signs of abnormality among toddlers?
Which intervention works best in addressing a specific type of abnormal behavior among small children (e.g. inability to speak, inability to crawl, losing eye contact, inability to respond to sound)
Is there a relationship between frequency and duration of parent contact and child development?
How do mothers respond to signs of abnormality in their children?
3. Classroom and learning
There have been a considerable number of literature emphasizing the role of environment in facilitating brain development. Intelligence cannot just be attributed to genes but also to the surrounding environment. In educating children, the learning environment can play a very significant role in shaping the minds of growing children. Classroom design is one such concern.
Example Research Questions
General Question: Does classroom design affect children behavior?
What are the development needs of children that can be addressed by classroom design?
How can classroom designs be made more stimulating to children?
Is there a relationship between classroom design and the number of uneasy children?
How do placements of play structures influence children behavior?
4. Socialization among children
With the preponderance of the internet nowadays, children become more engrossed in their laptops or desktop computers as part of their socialization activities. They now have more friends in the internet than in real life. Teenagers, in particular, are beset by this problem. Since they spend most of their time in front of computer screens, they were referred to as the “screenagers.”
There have been increased concern on children being unable to socialize effectively with real persons as a result of too much “socialization” activity online. This also exposes them to grave abuse as a result of their online ventures.
Example Research Questions
Is there a relationship between the frequency and duration of online activity among children and study time?
Is there relationship between age and the type of online material visited by children?
How do children respond to strangers requesting details about them?
Is there a relationship between children’s irritability and intensity of computer use?
What factors influence the time devoted by children to online activity?
These topics will enable you to get started on your research endeavor along the field of education. If you want to explore other topics, you might want to try other means such as brainstorming. This will help you zero in on topics that you might find interesting.
Child Development Media, Inc., n.d. Videos, books and curricula for child development professionals and parents. Retrieved on July 30, 2013 from http://www.childdevelopmentmedia.com/child-development-articles/