What is Quantitative Research (Tutorial 2019)
Qualitative research can be contrasted with the quantitative research paradigm that has dominated psychology along several dimensions. Qualitative research is not directed to a single objective reality; rather, the focus is on how people invest their experiences with meaning, and the assumption is that such subjective interpretations will vary across individuals.
Because people vary, the emphasis in sampling is less on representativeness and generalization and more on identifying populations that are of interest in themselves. This tutorial explains the Qualitative research and its Characteristics with best examples.
The goal in studying such populations is not to test theoretical predictions or to make group-level comparisons; rather it is to understand their subjective experiences, working as far as possible from the perspectives of the group members themselves.
Finally, numbers and statistical comparisons are not intrinsic to the qualitative approach; the concern rather is with a rich description of people’s experiences.
The goals of qualitative research are reflected in the methods that are used. In general, the quantitative researcher’s emphasis on control and manipulation is replaced in qualitative research by a concern with context and naturalness and exploration rather than hypothesis testing.
More specifically, seven general qualitative methodologies can be identified: narrative, phenomenology, grounded theory, discourse analysis, focus groups, ethnography, and case study.
Of these, the text elaborates on narrative research, ethnography, and case study, giving examples of developmental study in each case.
The blog concludes with three general points about the qualitative approach. The first point concerns the diversity in writings about the approach; although a common core of agreement exists, different authors may differ in the distinctions they draw and the forms of research they emphasize.
The second point concerns the status of the approach in contemporary developmental psychology. The conclusion is that the qualitative approach is a growing but still a minority position.
Finally, the third point is a reiteration of the value of converging operations. It emphasizes the complementary nature of the qualitative and quantitative perspectives and the desirability of basing general conclusions about a content area on both.
Characteristics of Qualitative Research
A reasonable way to introduce what is meant by qualitative research is by contrasting the qualitative approach with other approaches to research. The main contrast, as you might guess, is with so-called quantitative research. It is quantitative research that has been our main subject at this point in the blog.
Several features distinguish quantitative research. First, as the name indicates, quantitative research is quantitative—it involves attaching numbers to the units measured and performing statistical analyses on the numbers.
A further feature of quantitative research is an emphasis on group-level comparisons. Responses of specific individuals are seldom a concern in the quantitative approach; indeed, the attempt is often to reduce the “error variance” associated with individual differences among people.
Participants within a group (that is, a particular level of an independent variable) are assumed to be the same, and the focus is on possible differences between groups.
A third characteristic of the quantitative approach is an emphasis on representativeness and generalization. This point ties into the preceding one.
Individual participants in a quantitative study are not of interest in themselves; they are of interest to the extent that they tell us something about some broader population of interest. Issues of representative sampling are therefore critical.
A fourth characteristic is an emphasis on causal explanation. This emphasis on causality is reflected in the priority accorded to internal validity: the accuracy of cause-and-effect conclusions concerning how variables relate. Methodologically, it means a preference for experimental manipulation of variables when possible.
A final characteristic concerns the typical starting point for such research. The typical starting point is some basic theoretical question or set of questions that the research literature suggests is both important to answer and not yet completely answered.
The goal of the research is then to provide the research community with further scientific knowledge. The emphasis is on an objective value-free search for what is assumed to be objective reality.
How, then, does qualitative research differ? Let us begin with the final point concerning the starting point for research. The typical starting point for qualitative research is not the search for some aspect of objective reality that is the same for everyone. A basic assumption, in fact, is that such a search is misguided.
There may (depending on one’s philosophical preferences) be a constant and knowable physical world to which the natural sciences can speak. The target for the social sciences, however, is not the world itself but people’s experiences of the world, and such experiences are variable, personal, and self-constructed.
It is the meanings with which people invest their experiences that should be our concern, according to the qualitative perspective, and these meanings may vary from person to person.
The quantitative researcher’s search for an invariant objective reality is thus replaced by the qualitative researcher’s search for variable forms of subjective meaning.
The focus on people’s interpretation of reality has implications for issues of sampling. Because people’s experiences are in part personal and subjective, there is no expectation that one person or group will necessarily be representative of some other person or group.
Samples tend to be studied, therefore, not for their representativeness but because they are of interest in themselves.
Often they are of interest because they are somehow marginalized or neglected or at risk—for example, children growing up in poverty, or adolescents with an emerging gay or lesbian sexual orientation, or women who face barriers and discrimination in the workplace.
Issues of values are often an important component of qualitative research, and some (although by no means all) such research has an explicit political agenda.
If our concern is with how people interpret their worlds, it would make little sense to study them in tightly controlled and artificial laboratory environments. The quantitative researcher’s emphasis on control and manipulation and causality gives way in qualitative research to a concern with context and naturalness and rich description of people’s experiences.
The one-way-expert-to-subject orientation of quantitative research gives way as well to a more egalitarian interdependent relationship between researcher and research participant. The researcher needs to enter into the participant’s world in order to understand it, while all the time remaining aware of the difficulties in fully doing so.
A final contrast is, of course, the one conveyed in the names for the two approaches. Qualitative research is not necessarily number-free; some qualitative projects do include numerical tallies and perhaps some statistical comparisons.
Numbers are not intrinsic to the approach, however, and any quantitative component is simply a step en route to the final goal—namely, to capture how people make sense of their worlds.
Rather than numbers, therefore, primary data typically take the form of words, both the participants’ own words and the investigator’s interpretation of those words and related behaviors.
Methods of Study
How do qualitative researchers collect their data? In part, in the same way as do quantitative researchers. There are, after all, only so many ways to gather information about what people are like. Observations of naturally occurring behavior play an important role. So do interviews with the samples of interest.
It is true that some techniques in the quantitative researcher’s arsenal are unlikely to be found in qualitative studies—in particular, any that involve the automatic or essentially automatic recording of behavior (e.g., physiological measures).
Beyond this difference, what distinguishes qualitative from quantitative is less the specific technique of data collection than the approach taken to it.
The qualitative researcher who opts for an interview approach is unlikely to use a closed-choice interview or questionnaire on which the respondent chooses among options provided by the experimenter.
Much more likely is the use of an open-ended form of questioning, with the direction of the questioning determined at least in part by how the participant responds. Similarly, the qualitative researcher who decides to observe behavior is unlikely to employ a preset coding system in which boxes are checked to record the behaviors under study.
More likely is some form of running record that is open enough to capture anything of interest, with, again, the possibility of a change in what is recorded as the study develops.
Of course, whatever the method of data collection and the nature of the data, the final step is always the attempt by some human observer to make sense of what has been found.
In quantitative research, however, it seems fair to say that the goal is to minimize this interpretive aspect—to produce results that are so clear and so objective that anyone will interpret them in the same way. In qualitative research, the judgments of the individual researcher play a larger, indeed intrinsic, role.
As the preceding discussion suggests, standardization does not occupy the honored position in qualitative research that it does in quantitative. If standardization does enter in, it is often late in the data-gathering process rather than as an essential element from the start.
A related point is that much of what is done under the qualitative heading corresponds to what was discussed as exploratory research in Next Blog.
Quantitative researchers, of course, also do exploratory research, but such efforts do not occupy a very high proportion of most quantitative researchers’ time, and they typically are simply a prelude to more systematic and standardized study. In qualitative research, exploration is often the entire story and not simply the prelude.
These general points still leave unanswered the question of exactly how qualitative researchers go about doing their research. The specific methods that are used in qualitative studies can be organized in various ways. The listing is not exhaustive, but it does capture the most general and important techniques.
In what follows I elaborate on and give examples from the approaches that have arguably been most important so far for developmental psychology: narrative research, ethnographic research, and case study research.
As the name suggests, narrative research focuses on narration—the stories people tell about their lives. In Murray’s words, “Narrative psychology is concerned with the structure, content, and function of the stories that we tell each other and ourselves in social interaction.”
Because even quite young children can and do tell stories, it is a method well suited for developmental issues.
The narrative approach has been employed with a variety of different populations, ranging in age from toddlerhood through old age and spanning dozens of different cultures or subgroups. It is, in fact, one of the primary methods of study within the discipline of cultural psychology.
The method for collecting narratives vary some depending on the population being studied and the goals of the researcher. Sometimes narratives may be observable in people’s spontaneous interactions, sometimes they may emerge in informal conversations between researcher and participant, and sometimes they may be elicited through more formal interviews.
Although narratives may be elicited as part of the research, it is important to emphasize that they are not simply a research tool for learning about people. Rather, producing narratives is something that people do spontaneously, quite apart from the demands of research.
Narratives are a basic way in which we come to understand both ourselves and those around us. They are also one of the ways in which adults socialize children.
[Note: You can free download the complete Office 365 and Office 2019 com setup Guide for here]
Case Study Research
As the term suggests, case study research involves the intensive study of an individual case that is in some way informative enough to merit concentrated attention.
Beyond this general commonality, case studies can take a variety of forms. To begin with, not all case studies fall under the heading of qualitative research, for case studies can also be carried out from a quantitative perspective.
Whether the focus is qualitative or quantitative, a particular case may be selected because it is of interest and importance in itself (which is often true in clinical or educational case studies), because it is believed to be reflective of some larger issue, or, of course, for both reasons.
To the extent that some larger issue underlies the research, a case may be selected because it is believed to be typical of the situation that the researcher wishes to understand.
Often, however, a case may be selected precisely because it is not typical but rather is in some way atypical or extreme, and thus provides forms of evidence that are not available if we sample only within the normal range of experience and development.
The methods employed to analyze the case may also vary. Indeed, any or all of the techniques found in the qualitative approach in general—observations, interviews, written or archival records, textual analysis—may be employed in cases studies as well.
As one proponent of the approach writes, case study research offers a palette of methods. There are many, many ways to do case studies
A final distinction concerns the unit of study. In some instances, the unit may be a single individual. Perhaps the most famous case study in child psychology, that of a little girl who was dubbed Genie in reports of the case, falls in this category.
The case studies that are the subject of the final exercise at the end of the blog are a further example.
Often in the case of study research, however, the unit is not an individual but some larger group or program or institution. Thus, we might carry out a case study analysis of a school that is undergoing desegregation, of recently formed single-gender classrooms, of older adults who have returned to school, or of the spouses of individuals with Alzheimer’s.
Quantitative research designs include various methods including surveys, experiments, and content analysis.
Since the most commonly used quantitative research method is a questionnaire survey, we will focus on the steps involved in survey research and assume the discussion will help those interested in other quantitative research methods as well. Students frequently ask us these questions when designing their survey research projects:
What are my independent variable and dependent variables?
How do I select a sample to study from my target population?
What is an acceptable sample size for survey research?
How do I turn my concepts into variables in the survey questionnaire?
What are the levels of measurement and why do they matter?
The first half of this blog responds to these questions and relevant issues. In designing a survey research, the following steps are usually necessary:
What Are Your Independent and Dependent Variables?
The term “independent variable” is commonly used in social sciences to refer to the cause or the variable that affects the other in a hypothesized relationship.
The term “dependent variable” refers to the effects or outcomes in a hypothesized relationship. For example, let’s consider the research question, “How do relationships with parents affect teenagers’ school performance?”
Suppose you expect that teenagers who do not have the typical quarrelsome relationships with their parents will do better in school than those who have a lot of conflicts with their parents. The independent and dependent variables are already implied by your research questions.
The independent variable in this example would be relationships with parents and the dependent variable would be school performance.
Similarly, if your research question is “Are teenagers’ grades negatively affected by gravitation toward social media?” then your independent variable is “gravitation toward social media” and your dependent variable is “grades.”
Since you are likely to have more than one research questions in your study, you may have multiple independent and dependent variables.
Sometimes, you may have several independent variables and one dependent variable, and vice versa.
For example, questions such as “Do regular medical check-ups, exercise, and sufficient vegetable intakes reduce the likelihood of cancer?” and “Do cigarette bans in public buildings and higher cigarette taxes encourage smokers to quit smoking?” have multiple independent variables and a single dependent variable.
On the other hand, a research question on the academic and emotional effects of bedtime reading during early childhood assumes one independent variable (“bedtime reading”) and multiple dependent variables (the various “academic and emotional effects”).
It is a good practice to write down your research questions and label your independent and dependent variables.
When you identify and label your independent and dependent variables, you should be quite clear in your mind that an independent variable is the cause of the dependent variable and a dependent variable is the effect of the independent variable.
A dependent variable must be able to vary or be affected when it is influenced by the independent variable.
In another example, if you use education as an independent variable and salary as a dependent variable, then, you are anticipating that the salary of your respondents will change when their level of education changes.
If a variable cannot vary or cannot be affected, then it cannot be used as a dependent variable. For example, someone’s race and gender cannot be changed by the influence of other variables; thus, they cannot be used as dependent variables.
In the examples above, abstract concepts such as “relationship with parents,” “gravitation toward social media,” and “emotional effects,” need to be more specified and operationalized into measurable indicators so that you can quantify them.
Operationalization is a step where you identify very specific indicators or measures for your concepts. For example, “relationships with parents” are not something you can directly observe, but you can use some very specific indicators for a good or a bad relationship.
Quantifiable indicators, such as “number of arguments a teenager had with his/her parents within a month,” “number of times a teenager received a punishment from parents within a month,” and “number of times a teenager violated rules set by parents” are all good ways to measure whether a teen has a good relationship with his/her parents.
Or, you can simply ask the teen respondents to rate the quality of their relationship with parents on a scale of one to ten. The concept, “gravitation toward social media” is hard to measure itself. You will need to use tangible measures such as “time spent on social media each night.”
Likewise, the concept “emotional effects” can be specified into multiple questions gauging how happy the child is, how social the child is, or how energetic and curious the child is, and so on.
How Do You Select a Sample to Study from Your Target Population?
What group of people or cases is your research about? Do your research questions concern the general population, a particular group of people, countries, schools, or other social organizations?
The answer to these questions will be your study population or target population; the term refers to the group of people or cases about whom you will conduct your study and to whom you will apply the findings of your study. Your population is also the pool of cases from which you will select a sample or a subgroup of the cases you will actually study.
As you can imagine, if you select a sample that resembles your population closely, you will be able to use your findings to tell something about your study population.
But if your sample does not resemble your study population, your ability to use your study findings to predict the patterns in the study population is limited. Suppose you have selected a group of students from your university whose average grade is an A.
You know it is unlikely that this sample will reflect what the average grade is in your university.
The extent to which your sample “looks like” your study population is called “representativeness”; study findings from a representative sample can be generalized to the study population.
For instance, if a group of spectators selected by random drawing of numbers happens to have the same demographic characteristics as the spectators in the entire stadium, this sample will be representative of the crowd in the stadium.
This means that, if there is more rooting for Team A in this sample, you can generalize that there will be more support for Team A among the entire stadium crowd.
How do we select a representative sample? Social science methods teach us that we can approximate a representative sample by reducing systemic selection biases in the sample drawing process. In general, a selection method which only relies on random chance is considered as having no systemic selection biases.
There are a variety of different ways to draw a sample from the study population: simple random sampling, systematic random sampling, stratified random sampling, cluster sampling, quota sampling, snowball sampling, purposive sampling, and availability sampling. Some of these sampling techniques select participants using random drawing while others do not.
The specific steps and details of different sampling strategies are beyond the scope of this blog. If you need to refresh your memory on how to draw a particular type of sample, consult some of the references listed in this blog.
In the broadest sense, sampling methods fall into two groups: probability and non-probably sampling. In the above list, all the variety of random sampling and cluster sampling fall in the probability sampling category.
Quota sampling, snowball sampling, purposive sampling, and availability sampling are non-probability sampling. Probability sampling methods select participants based only on random chance. Sampling theory considers this the best way to obtain a representative sample.
To use probability sampling methods, you need access to the sampling frame, which refers to the roster of all units in your study population.
e.g., an approximate list of citizens of a country, a list of residents of a community, a list of all schools, organizations, student roster, and so on.
So that everyone is in the pool of available subjects and only random chance can determine whether someone is selected to be included in the sample.
Non-probability samples are used when researchers do not have access to the sampling frame or do not have a clearly identifiable study population (such as undocumented immigrants, homeless population).
The research was done by students like you often has to resort to non-probability sampling methods simply due to insufficient time and resources. Non-probability sampling methods are likely to introduce sampling biases because factors other than random chance will affect the selection process.
It is okay to study a non-probability sample, especially for a small-scale exploratory study, or if you are conducting qualitative research. Just keep in mind that your findings will have limited generalizability, and the limitation should be included in the discussion of your findings.
When you select your study population, make sure that you can gain access to them.
If your study population is minors (such as children or juveniles) or people with limited power (such as prisoners), you may face particular difficulties obtaining informed consents from guardians or getting permission from the heads of the institutions to enter the sites to collect data for your research.
Therefore, think carefully about access before you decide to study a particular population.
After you decide on your study population, decide on what your unit of analysis is. It may be individuals, universities, organizations, or countries, depending on what is most appropriate for your research.
What Is an Acceptable Sample Size for Surveys?
Another issue to consider is sample size. Regardless of whether you use probability sampling or non-probability sampling, the size of the sample is an independent issue which requires your attention.
If you want to conduct surveys and use computer software to do data analysis, you need a sufficient number of respondents in your sample. As a general guideline, a minimum of 400 cases will be amenable for statistical data analysis. This suggestion is to reduce sampling error due to sample size.
If you have a sample size of 400 cases, the standard sampling error will always be 5% or smaller no matter what the variation is in the study population.
In reality, however, it may be unrealistic for a student researcher to be able to draw a sufficient size sample; you are more likely to work with much smaller size sample due to the time and resource constraints.
In deciding your sample size, consult with your project supervisor, or professor, as they may have specific guidelines or requirements for sample size. Generally speaking, three principles are useful in determining sample size.
First, the larger the sample size, the smaller your standard sampling error will be. At the same time, your sample is more likely to resemble the characteristics of your population and you will be able to generalize your findings to the target population.
Second, if you are conducting a quantitative study, most statistical analysis techniques used in social sciences assume a normal or the bell-curve distribution of data.
If the sample is too small, say less than 100 cases, there is a good chance that you will not meet this assumption of a normal distribution. Our advice is to obtain at least a sample size of 100 respondents if you plan to use statistical analysis techniques.
With that number, you will be able to use commonly used descriptive and inferential statistical techniques such as cross-tabulation analyses, chi-square tests, t-tests for comparison of means, and so on.
Keep in mind that you may receive invalid answers, which you will exclude from your analysis; to obtain a sample size of a 100 valid cases; you may need to go slightly beyond your targeted sample size when you collect the data.
Third, the small sample size may produce insignificant statistical results simply as a function of the sample size. Sometimes, small samples require you to use special statistical measures other than the commonly used measures mentioned above.
According to probability theory, when the sample size decreases, the standard error increases. For example, if you do a chi-square test with a very small sample, you may find that many of the cells in your cross-tabulation have fewer than five cases and your chi-square value is not statistically significant.
If you have more than 25% of the cells with fewer than five counts, your chi-square analysis is not acceptable (George and Mallery 2000). If this is the case, you cannot use chi-square analysis to test whether two variables are statistically independent of each other.
On the other hand, if you only intend to use simpler descriptive statistics such as percentages and graphs to answer your research questions, a sample size smaller than 100 can still work.
A “robust” sample, or a sufficiently large and representative sample, is needed if your goal is to perform an explanatory study. This is one reason why many researchers turn to secondary data, data collected by governments or large organizations with resources, for explanatory studies.
As a student researcher, you are likely to conduct a study on a modest size sample using non-probability sampling strategy.
This is quite all right, as long as you are mindful of the limitations of your methods, make cautious interpretation of your findings, and discuss the limitations in your report.
If you and your supervisor decide to work with a very small sized sample, say less than 50 respondents, you may consider it as a pilot test for a full-scale study at a later time.
A pilot study can yield valuable information; you can get a sense of how survey questions are interpreted by the respondents, gauge whether the length is appropriate, estimate what response rates you can anticipate, and detect any potential problems in administering the surveys.
These are important learning experiences and skills-building processes which make your project meaningful. You will also obtain some descriptive statistics about the patterns of behaviors and attitudes you are investigating with a pilot study.
If you must resort to very small sample sizes due to the constraints in time and resources, consult your project supervisor about designing a well-planned pilot study which will give you a good foundation for future projects; this can be more meaningful assignment than a poorly conducted research on a larger sample.
Also, the above guidelines are for quantitative studies requiring statistical data analyses. Qualitative research is most often done on a sample much smaller than Sampling issues in qualitative research will be further discussed later in the blog when we review steps of qualitative research.
How Do You Turn Your Concepts into Variables in Surveys?
If you conduct a survey, develop your survey questionnaire to include questions on the independent and dependent variables and demographic information needed for your research.
Most surveys include demographic questions to collect basic information about your study population. Typically, you ask about gender, age, level of education, and socioeconomic status, and race and ethnicity. However, which information is needed depends on your research questions.
For example, if you study university students’ adjustments to campus life away from home, it may be relevant to ask how many years they have attended university. Before you start to construct your survey questions, you should have a clearly identified list of your independent, dependent, and demographic variables.
If you have an abstract concept in your research question, In surveys, most concepts are measured by one or a set of questions.
For example, if your research question is “How do gender, race, and age affect student school performance,” then you will need to specify and measure school performance.
In this case, you cannot directly ask your respondents what their school performances are. Instead, you need to identify indicators such as student Grade Point Average, class attendance, or time spent on a study to measure school performances.
Since you need to include measures for all your concepts, you should have at least one question for each of your concepts in your survey instrument. Some simple variables such as gender, age, whether or not one is in university, whether or not one supports a political candidate can be sufficiently measured by a single question item.
But for abstract concepts or concepts that have a broader range of meaning, you may need more than one indicator.
For example, if your variable is depression, a composite question consisting of several items asking about different symptoms of depression may be a far better measure than a single question asking whether one experiences depression.
The key idea is that the indicators in your survey instrument should capture the full meaning of your concepts.
In the U.S., students in university classes participate in teaching evaluation surveys at the end of the semester. The purpose of the surveys is obviously to measure “good teaching.”
Imagine that schools use just one question, “does your professor demonstrate good teaching?” This single item indicator would be not only too simplistic but also vague as each student may have different ideas about what good teaching is.
Typical teaching evaluation surveys thus include several questions on various aspects of teaching, such as knowledgeability, organized presentation, use of examples, interaction with students and so on.
In developing survey questions, please pay attention to the validity of your questions to make sure that you will get what you intended to get, or measure what you mean to measure.
For example, student GPA is a good indicator of academic performance. How many times students visit the library may not be a direct measure of academic performance.
It may affect academic performance, but library use itself is not a valid measure of academic performance. Your measures should also have reliability which refers to the quality of a measure that produces the same value or observation repeatedly.
For example, reading a thermometer is a much more reliable measure for fever than feeling the temperature with your hand on someone’s forehead, since the judgment from touching one’s forehead can be capricious. Likewise, asking “how many times did you drink last week?” is a more reliable measure than asking “do you drink often?”
What Are Levels of Measurement and Why Do They Matter?
When you design your survey questions, think about what kind of data analysis you are going to conduct after you collect your data. Statistical data analysis procedures are often closely related to the mathematical property of your variables, called “the levels of measurement.”
Given that statistical analyses are based on mathematical computations of various statistics, the mathematical qualities of the numerical data you collect will influence data analysis.
Let us use a simple example. For the demographic variable, gender, researchers often give the numeric code “1” for male respondents and code “2” for female respondents.
Computing what percentage of your sample is men or women would be an appropriate thing to do, but computing the mean score for this variable is not, as there is no such thing called “average gender.”
You would not want to use the statistical technique “mean score” when you perform an analysis of variables such as religion, nationality, or favorite Thai dish. This is because all of these examples are variables with no mathematical meaning or no variations in amount.
On the other extreme, if you ask respondents to write down their age, or their salaries in precise currency terms, then you know that the different values you will obtain can be compared in quantity; computation of mean scores (e.g., average age, average income) makes perfect sense with these variables.
These examples illustrate the different mathematical qualities of the numeric data you will obtain from different types of variables.
As we mentioned above, statistical tests assume that the variables entered into the computation have a certain type of mathematical property. The types of mathematical properties are expressed in levels of measurement.
In statistical analysis, four levels of measurement are differentiated, depending on the level of mathematical precision of the measure: they are nominal, ordinal, interval and ratio levels.
A nominal level of measurement has the lowest level of mathematical quality because its values are number codes that do not represent actual quantities, but simply different qualities.
A ratio level of measurement, on the other hand, is the highest level of mathematical precision because its values actually correspond to mathematical quantities, just like natural numbers. Let us explain each of these with some examples.
Nominal level measurements pertain to those variables whose attributes are simply different in type, not in quantity, from one another. In other words, the numeric codes assigned to the respondents’ different answers for a nominal level measure have no implied order or quantity.
The difference between categories of a nominal scale is a qualitative difference. For example, gender, race, occupation, and marital status are usually measured at a nominal level, as different attributes respondents can take in these variables are different kinds and not different amount.
Ordinal level measurements assess a variable in its order of magnitude. They have the property of magnitude or a rank-order.
In other words, variables that are measured at the ordinal level have categorical attributes like those of nominal level measures, but they also have the advantage that the categories can be ordered or ranked from low to high or from one extreme to the other.
Thus, the numbers assigned to different attributes, or categories signify different point in a system of order. For example, social class, people’s attitudes, or students’ academic standing are usually measured at the ordinal level.
The interval level of measurement measures a variable’s attributes that are not only rank-ordered but also are separated by a uniform distance between them.
Interval level variables are measured in terms of a standard unit of measurement, and therefore the categories of an interval measurement have equal intervals between the categories. For example, the temperature can be measured at an interval level.
Question: What is the temperature in your city today? ______ C°
In this example, the measurement indicates an equal difference between any two units of the measurement. That is, the difference between 18C° and 19C° is the same as the difference between 10C° and 11C°.
But notice that there is no absolute zero, or absence of a value in this scale. A temperature of 0C° is still a temperature indicating a relative location in this scale, and it does not mean that there is no temperature.
Ratio level of measurement is the same as interval level of measurement, except that the ratio level of measurement is also based on a true zero point or an absence of the quantity. In other words, variables that are measured on a ratio scale have the characteristics of an interval scale plus a real zero. For example:
Question: How much money do you have in your pocket now? ________ Euros
In this example, if a respondent answered zero, he/she has absolutely no money at all in his/her pocket. Likewise, if someone answered zero, for a question about years of schooling, you can interpret this as a lack of any formal schooling.
In social science research, you will find very few, if any, true interval level variables; most measures are ratio level measures. Also, the distinction between interval level and ratio level measures does not make a difference in statistical computations for social sciences. Therefore, you may treat the interval and ratio level of measurement as if they are the same level.
This is a common practice and the reason why many people lump the two together as “interval-ratio” level measures.
If you use software packages for statistical analysis such as IBM’s SPSS program, you will notice that there is no option for designating a variable as an interval-level measurement. Instead, interval and ratio levels of measurement are both indicated with the term “scale.”
Since the levels of the measure will enable or limit the statistical procedures you can apply at a later stage, it is important to understand different levels of measurement at this stage and operationalize your variables into adequate levels of measurement.
For instance, you can use bar charts or pie charts, frequencies and percentages, and cross-tabulations to analyze your data when your independent and dependent variables are a nominal level of measurement.
If you want to calculate means, correlations, or multiple regressions, however, your independent and dependent variables need to be interval or ratio level measures.
Similarly, to use means or variances to compare two groups (t-test) or several groups (one-way analysis of variance), your dependent variables should be measured at the interval or ratio level.
If you are not sure about what kind of data analysis you will eventually use at the time you design your survey questionnaire, we advise you measure your independent and dependent variables at the interval or ratio level whenever possible.
Of course, some variables, such as gender, ethnicity, and religious affiliation, are naturally nominal level and cannot be measured as ratio level variables. But there are many variables that can be measured on an ordinal scale or a ratio scale;
for example, salaries from work can be measured as an ordinal level variable (“Which of the following range does your monthly salary belong to?”) or as a ratio level variable (“Please write down the exact amount of your monthly salary in U.S. dollars.”).
In this case, we recommend that you use a ratio level variable. We suggest this because statistical software packages allow you to transform ratio level variables into a lower level measurement after the data are entered, but you cannot do the reverse.
For example, if you collected your income data in real dollar amount (i.e., a ratio level measure) but later decide to use the simple cross-tabulation technique for analysis, you can recode the income variable into a new four- or five-category income range variable (i.e. an ordinal level measure).
No information is lost in this recoding process, and you can still keep the data on your original variable (the exact dollar amount for each respondent) as a separate variable.
If your original variable was an ordinal level income range variable, however, there is no way you can estimate what exact dollar amount each respondent earned. The lesson here is that mathematically higher level measures tend to contain more minute information.
Therefore, you should measure at the highest level possible because you may need information with more precision at the data analysis stage.
Similarly, it is a good idea to use ordinal level measures when you have an option to choose either nominal level or ordinal level. For example, students often construct questions with simple “yes” or “no” answer choices.
You may ask yes/no questions but after collecting data, realize that the procedures you can use to analyze data are quite limited. Yes/no questions contain far less information than asking, for instance, about the frequency of particular behaviors or the degree of support on specific issues.
The lesson is that, in designing your questionnaire, consider what kind of data analysis you will conduct and determine the appropriate level of measurement for each of your variables.
Quantitative research designs are suitable to pursue relatively straightforward questions and clearly defined hypotheses.
If your research questions call for the in-depth understanding of a complex problem, a deeper examination of a particular experience, or comprehensive knowledge of a group of people, you will be better off collecting detailed qualitative data for your empirical research. We now discuss some methodological issues relevant to qualitative research.
There are more issues related to survey designs which we have not covered in this blog. Some are beyond the scope of this blog.
We hope that you will find more in-depth details from your research methods textbooks. One issue we have not discussed in this blog is how to write clear and effective questions. There are many tips for properly wording survey questions.
What Do You Need to Know about Qualitative Research Designs?
The main benefit of qualitative data collection methods is two-fold: 1) they allow you to obtain data reflecting the participants’ own perspectives, and 2) they are conducive for a wide range of observations without a pre-conceived notion about the issue or the situation.
These methods are designed to have participants’ voluntary information in their own words or let you make observations in the natural setting of a situation.
There are various ways of conducting qualitative research, and each way may be appropriate for a specific type of research project. some of the popular qualitative data collection strategies include in-depth interviews, participant observations, ethnography, and focus groups.
Since in-depth interviews are most frequently used by students, we will focus on the steps and concerns involved in in-depth interviews. When designing their qualitative research project, students frequently ask us these questions:
How do I construct my interview questions?
How do I select people for interviews?
What should I do during the interviews?
What other qualitative data collection methods can I consider?
The remaining part of this blog responds to these questions and relevant issues. In designing in-depth interviews, the following steps are usually necessary:
How Do You Construct Your Interview Questions?
In-depth interviews are based on open-ended questions that reject pre-conceived predictions and allow participants to express themselves in their own words.
They offer you a better sense of what respondents feel and think. For example, suppose you are interested in understanding the challenges new immigrants face in adjusting to their host country, your research question may ask:
What are the main problems new immigrants and refugees face in integrating into their receiving country? This question is broad and open-ended. You have no independent or dependent variables, nor is there a need to identify them.
But you are not approaching this question blindly, either. Having reviewed existing literature, you probably now know something about issues that concern new immigrants.
For example, you know that the literature on immigrants and refugees discusses legal barriers for work permits and citizenship, cultural differences, language barriers, prejudice and discrimination based on race, ethnicity, and religion, and economic difficulties.
With this knowledge, you can identify focal areas in designing your interview research and constructing interview questions.
In-depth interviews are not usually highly structured and allow you to spontaneously explore emerging new issues and themes as the interview unfolds.
To collect relevant information consistently across many interviewees, however, you need to have some standard set of broad questions you want to ask everyone. This set of questions is called an “interview schedule.”
Depending on how many specific questions you have in your interview schedule, and how closely you plan to adhere to the prepared list, you may have highly structured interviews, semi-structured interviews, or unstructured interviews. Which one you choose to do depends on the goals of your study. Here are a few tips for constructing an interview schedule.
First, keep the questions open-ended and allow your respondents to answer them in their own way. Questions leading to a yes/no answer are not productive in interviews.
For example, if you ask, “Did you have a lot of difficulties finding jobs after you came to this country?” your interviewee is likely to simply answer, “Yes, I did” and stop.
Instead, if you ask, “Tell me about how you tried to find work when you arrived here, and how you found your first job,” you are inviting your respondent to tell you about his experiences and difficulties with his job search.
Some immigrants may not have had much trouble finding a job because a relative had a job offer already lined up while others lacking connections may have had a long period of unemployment and economic hardship.
Respondents may voluntarily tell you stories about demanding and unrewarding work or share their experiences of discrimination while telling their job search stories.
You will be in a better position to collect these varying accounts on work when you use open-ended phrases like the second example.
Moreover, you will be able to assess what issues matter most about work and job opportunities by paying attention to the issues your interviewees mention first. These rich details will not surface if you ask “yes/no” questions.
Second, for each interview question, you should think about “probing questions,” or follow-up questions for more details or information on a closely related issue. Probing is a critical skill in interviews since your respondents are unlikely to pour out details about their experiences right from the beginning.
In our example about the job experience, your interviewee may give you a relatively short answer, for instance, “I have a friend who worked in a restaurant. He told me that his manager may give me a job.
So, I went with him one day and started to work there.” At this point, probe further in a few different directions.
If your research focus is job market conditions and opportunities for immigrants, ask follow-up questions about how long he searched before getting his first job, what he did to search for work, what types of the job were available, and so on.
If your research question concerns the role of social networks for new immigrants, which many scholars consider important, then probe more about this “friend” who helped your interviewee. Ask: “Who was that friend?
Was he someone from your country? Did you meet him here?” These questions will help your respondent elaborate on his relationship with his friend and possibly other resources and information that may be vital to his settlement into a new community.
Third, keep your interview schedule flexible enough so that you can pursue new leads or switch the order of your questions as the conversation unfolds. Depending on your research questions and style, interviews can be more or less structured.
You may feel skilled and comfortable conducting interviews only with an outline of themes in mind (unstructured interviews).
Or you may be concerned about forgetting some important questions and would prefer to stay close to the prepared interview schedules (structured interviews).
Most of you are likely to be somewhere in-between and follow a style called “semi-structured” interviews. In semi-structured interviews, you will prepare an interview schedule including some probing questions but you will also let yourself explore a different theme or direction as new clues emerge in the conversation.
Be ready to embrace newly found issues in the course of the interview. In our example of an interview an immigrant, suppose your interviewee told you that one day he got into a fight with another immigrant worker from Africa because he did not like the African song this man was singing to himself all the time.
This may prompt you to ask about possible conflicts between workers of different ethno-religious backgrounds in places where many immigrants work.
Even if you had not planned to ask questions about this issue, this may well be a relevant focal point in your research. Do not hesitate to add new questions, if there are newly emerging interests as your interviews progress.
Fourth, interviews are conversations and the flow of the interviews may affect the amount of information you obtain. In general, it helps the interviewees’ thought process if you move from one question to the next with a sense of continuity.
For instance, ask about marriage and then about children; there is a close connection between these themes and your interviewee is likely to stay focused and remember more relevant details.
But if you ask about marriage first, then educational background and old school friends, and finally about children, it may be difficult for your respondent to recall information, jumping back and forth on different periods of the life course. This consideration of smooth flow can guide you in constructing your interview schedules as well as carrying out the interviews.
Overall, having a prepared interview schedule/guide offers at least two advantages. One is that you will have a clear direction before you start your interviews and will be able to include all your main questions.
The other is that you are likely to collect consistent information from all interviewees. As long as you use the interview guides with flexibility, you will enhance your chance of obtaining richer and more useful data.
How Do You Select People for Interviews?
Just as in quantitative research, qualitative studies require careful thoughts on sample selection. Selecting appropriate participants for your research will have an impact on your data since you are likely to have a relatively small sample in an in-depth qualitative study. It may be unrealistic to gather a sample that is truly representative of your study population.
One way to handle this limitation is to attempt to select a sample heterogeneous enough to capture the diversity of experience and perspective on the topic you are studying. Selecting interviewees parallels the process of sampling in quantitative research, except that you are highly likely to rely on non-probability sampling methods.
When you select participants for an interview-based study, carefully consider the heterogeneity of the sample, a minimum threshold of interviewees, and the pros and cons of different sampling methods.
For the heterogeneity of your sample; you probably want to select people of different demographic attributes so that you can collect accounts from diverse points of views.
For example, if you want to know about experiences of homeless people in a large city, include people of diverse demographic backgrounds, with dependent families and without, from different service facilities, and from different geographic sections of the city.
Beginning researchers often resort to interviewing those who are easily available. But if you select people solely in the same social network, you may obtain similar stories; people in the same social network tend to have common characteristics.
It is convenient to recruit participants from one support group for recovering alcoholics, for instance, but this is not the best strategy for discovering a range of patterns, experiences, and stories about this population.
To increase the heterogeneity of your sample, you may purposively reach out to participants with diverse social characteristics (e.g., by gender, age, nationality, ethnicity, level of education, race, and so on).
Selecting participants from multiple geographic locations is one strategy to increase diversity and improve representativeness within your sample.
Second, even if a relatively small size sample is acceptable for qualitative studies, you need to have a minimum number of participants in your sample.
If you interview only four or five individuals, you will be unable to find enough commonalities or draw any generalizable conclusions. In this case, the outcomes become little more than five individual stories.
If you are conducting a case study which is the type of qualitative study zooming into a comprehensive and in-depth examination of a single or a few cases, then five individual stories would still be good for case-oriented analyses.
But if your objective is to summarize some theoretical themes, then you may need a minimum number of interviews to represent a common story of your study population.
The sample size in qualitative research is usually not pre-determined, but a “saturation point” is often recommended as the threshold for concluding the data collection.
The “saturation point” is defined as the moment when additional interviews no longer produce new information; in other words, when you feel that you are getting the same story over and over, you may have reached the saturation point.
The saturation point can come earlier or later in your interviews depending on the scope of your research questions and the heterogeneity of your sample.
In our experiences, published qualitative studies tend to be based on more than 30 interviews and rarely go over a sample size of 100 unless secondary data sources were used.
For semester-long undergraduate research, we think a minimum of 13–15 interviews is realistic, although you may not be able to reach the saturation point. If you have a year or longer to work on your research project, you will have a better chance of getting enough interviews to reach the saturation point.
We strongly recommend that you consult with your project supervisors and faculty mentors who are familiar with the scope of your research questions and the constraints of resources under which you are conducting this research.
Since most qualitative research is exploratory in nature, a non-representative sample itself does not disqualify the entire research. You just need to interpret its findings with caution and avoid an overgeneralization.
A few non-probability sampling methods you may want to consider, instead of availability sampling, include purposive sampling, quota sampling, and snowball sampling. Purposive sampling involves selecting individuals who fit certain criteria required by your research questions.
Individuals may be selected because they belong to certain groups, demographic categories, or they are likely to have special information or knowledge that can help your study.
For example, for a study of a city’s sanitation services, you may purposively interview people involved in sanitation services including the city’s director, managers of various divisions, and street cleaning crew.
If it is important for you to include in your study various types of business organizations, you may also set a quota in your sample for various business categories and reach out to different business sectors according to your quota.
This strategy is called Quota sampling. Quota sampling helps you to ensure a representation of various groups of cases in your sample and to achieve a make-up of the sample that is similar to your target population.
Snowball Sampling is a strategy to recruit additional participants by utilizing referrals from earlier interviews. Just as a snowball grows in size by rolling it, you will rely on earlier interviewees to introduce new participants to you.
In snowball sampling, you will ask each interviewee to introduce you to other potential participants who meet the criteria to be included in the study; the chain of referrals will enable you to amass growing numbers of people into your study, like an enlarging snowball.
Snowball sampling is especially useful if your study population is not easily identifiable, such as homeless people, victims of intimate partner violence, or undocumented immigrants.
Although non-probability sampling methods are commonly used in qualitative research, this does not mean that you cannot use a probability sampling strategy. If your study population is a small sized group and if you have a list of everyone in the population, you can pursue the random sampling strategy described earlier in this blog.
For instance, if you are conducting a case study of a business corporation and you can obtain the list of all employees in this organization, make a random selection using the list.
But bear in mind that if the sample size is small, random sampling does not promise any greater representativeness than the non-probability sampling strategies listed above.
You may even use a combination of different selection strategies to recruit participants for your qualitative study. The key is to select a group of participants from which you will be able to collect a maximum amount of information for your research questions. Thus, the scope and goals of your research will determine the best methods for selecting interviewees.
What Should You Do to Have Productive Interviews?
Since interview research entails face-to-face interactions between the interviewer (you) and the interviewees, paying special attention to the following aspects of the interviewing process will positively affect the quality of your interviews.
First, you want to have a nice introduction. A proper and friendly introduction will set the tone of the entire interview. One of the main purposes of the introduction would be to create a comfortable connection with the interviewee.
There is another important purpose, which is to ensure the ethical process. During the introduction, you need to let the interviewee get to know you and learn about your study so that he/she feels comfortable in agreeing to participate.
Here are some very important things to include in your introduction: information about yourself (e.g., your name, institutional affiliation, position/title etc.), information about your study (e.g., the objectives, sponsorships of the study if any, the purpose of the interview, the use of data, etc.).
And information about the recruitment of the respondent (e.g., how his/her name was found, why he/she was selected, what makes him/her a suitable interviewee for the study, etc.), and what you will ask of the respondent (e.g., types of question to discuss, length of the interview).
Following the research ethics protocol, you should also clarify whether the interview will be tape-recorded, how confidentiality is guaranteed, and how you will safeguard the information collected.
If there is a potential emotional risk for the interviewee, you must disclose it before he/she agrees to sign the informed consent statement. Also, ask the interviewee if she has any other questions about the study.
Openness and respect are important guiding principles in this process. Remember that the participant is doing you a huge favor by agreeing to give you his/her time to share personal experiences and thoughts. At the same time, this favor puts her in a vulnerable position vis-à-vis you.
It is critically important that you should build a sense of trust in the first few minutes so that the interviewee can comfortably open up to you. Use thoughtful and non-judgmental language.
Sometimes, it will be necessary to avoid using certain words that may affect your interviewees. For example, you may avoid telling your interviewees that this research is about low self-esteem or drug abuse, which may make some interviewees uncomfortable or feel defensive. In such cases, you may use more neutral terms.
For example, you may use “substance use” instead of “substance abuse.” On the other hand, do not tell them anything that is untrue. Your use of the data collected from your interviewees must avoid hurting your interviewees, and you must tell them the purpose of your research and how you are going to use the data before the interview begins.
Second, interviewees may be nervous because, after all, you are a stranger, and they do not know what you will ask them.
Creating a comfortable and relaxed atmosphere for the conversation will enable your interviewee to remember better and to feel more inclined to give honest and more detailed stories, which will enhance the richness of the data collected.
Something as simple and mundane as dressing appropriately can affect the interaction.
If you dress too formally or too casually, it could become a cultural barrier to your conversation with your interviewees. You should try to talk in a similar style to your interviewee, using a level of vocabulary with which your interviewee will feel comfortable.
If an interviewee feels either inferior or superior to you, the conversation may be affected and become unproductive.
Ask questions in a conversational style. If you read questions as if you are reading from a blog, you may sound like you are testing your respondent rather than inviting him/her to talk. Practice your questions before you meet your interviewee so that you sound personable and welcoming in your questions.
Also, since interviewees often ask for clarification, you should be able to explain the questions in different ways. During the entire interview, you should maintain a genuine interest in the interviewee’s stories.
A good listener is the best encouragement for the story-teller. Keep in mind that your level of engagement with the interviewee’s story will affect how much information he/she will be willing to volunteer and share with you.
Smiling, nodding, short words of agreement, encouragement to go on, and taking notes are all good gestures to show your support and interest. Be careful about any subtle expressions or gestures you may give off, and make every effort to avoid value judgments. You should eliminate any personal biases before and during the interview;
for example, you should not assume that if your interviewee is from a poor neighborhood he/she will be unhappy or have low self-esteem. Likewise, if you say, for instance, “how do you feel about wearing hijab since it is a symbol of Muslim women’s subordination?” then your interviewee may feel that you are biased against her religion.
Third, your interview schedule may include questions regarding sensitive issues or traumatic experiences. In this case, we suggest you start the interview with more general and not-too-personal questions and move onto the sensitive questions later once a rapport is established.
For example, if your research is related to marriage, you may first ask your respondents general questions: when they got married, where and how they were married, and what their general perspectives on marriage are. Then move on to ask about problems and difficulties with their marriages.
Fourth, remember that the main purpose of your interviews is to gather information, and your primary task is to listen. Keep your talk to a minimum.
Say enough to maintain a comfortable conversational atmosphere but you should not talk more than your interviewee. In addition, be careful not to make statements or ask questions that will “lead” your interviewees.
What you want to get from your interviewees is accurate information, not something you like or with which you agree. People tend to agree, rather than to disagree, with the partners of interactions.
If your interviewee picks up clues about your own thoughts on the topic discussed, it could influence his/her responses. For example, during an interview about a national health insurance system, let’s suppose you asked,
“Given the problems with the long-wait and overcrowding of hospitals, what are your thoughts about this country’s national healthcare system?”
Because the way you mentioned those negative issues as given, the respondent will lean toward saying negative things. If this happens, the answers you obtain are not necessarily the respondent’s genuine thoughts.
Interviews essentially involve building a partnership with your respondent. Thus, the cultural expectations for respectful and professional social interactions should generally apply.
In addition, since your goal is to obtain truthful and valid information, you should make continuous efforts to make the interactional dynamics conducive to a focused and productive conversation.
But bear in mind that, ultimately, safeguarding your participant from any emotional or physical risks should be a prevailing priority in in-depth interviews.
Role of the Proposal
When it comes to research, very few projects get off the ground without some sort of approval. It may be as straightforward as verbal approval from your lecturer, but it is equally likely to involve a formal approval process gained through an admissions board, an ethics committee or a funding body. And of course, you may need approval from more than one of these.
This means you will need to develop a research proposal. Now many see the proposal as an opportunity to clarify thinking, bed down ideas and articulate thoughts in a way that will provide a study’s outline as well as a blueprint for future action.
And yes, a research proposal is all these things. But – and this is important – a proposal is not something you write for yourself. It is, without a doubt, a sales pitch. Your proposal is your opportunity, and sometimes your only opportunity, to sell your project and get your study off the ground.
So whether you are after admission to a university research programme, seeking ethics approval or looking for funding, the role of the proposal is to convince the powers that be that what you are proposing meets their requirements.
Namely, that the research question, the proposed methods, and the researcher all have merit. In other words, a committee will assess not only whether a project is useful and practical, but also whether or not it thinks you as the proposer have the ability to carry the project out.
Now keep in mind that the weight given to various aspects of a proposal varies according to the type of committee you are addressing and the type of approval you are seeking. For example, a proposal written to get into a Ph.D. programme really needs to sell your potential as a researcher.
A proposal written for an ethics committee needs to focus on the relationship between methods and participants. A proposal to a funding body, however, would need to have a strong emphasis on the practicalities of the method and the benefits of potential outcomes.
Demonstrating the Merits of the Research Question
Essential to any successful proposal is your ability to sell the merits of your research question. Demonstrating merits will rely on two things.
The first is that you are able to clearly and succinctly share your research topic and question (generally the work of the title, summary/abstract, aims/objectives, research question/hypothesis).
The second thing is that you can demonstrate that your research question is worth answering; that is, your question is significant enough to warrant support either at the level of admission to a programme or via funding (generally the work of the introduction/background/rationale).
When it comes to a committee’s assessment there are several possible scenarios:
The worth of the research question is self-evident (e.g. ‘What are the most effective strategies for curbing binge drinking in under-18s?’), and you are able to argue the importance and significance of your question to the satisfaction of the assessors. So far so good.
The worth of the research question is, as above, self-evident, but you do a lousy job arguing the case and do not convince the assessors that you are capable of mounting what should be a straightforward argument. Major problem.
The worth of the research question is not self-evident (e.g. ‘Do residents of the UK enjoy watching Big Brother more than US residents?’), but you are able to convincingly argue the case by citing evidence that attests to a real issue and what benefits there might be in conducting research into this area. If you can do this (particularly for this question), that’s impressive!
The worth of the research question is, as above, not self-evident, and you do little to help your case. Your arguments are weak so assessors are left scratching their heads and quickly put your proposal into the reject pile.
The point here is that while the significance of the research question is important, what is actually being assessed is your ability to argue the significance. It is therefore crucial that your writing be tight, well structured and well referenced.
Demonstrating Merits of the Proposed Methods
Once your assessors are convinced that your research question has merit, their focus will turn to methods. Here they are looking for several things:
Are the proposed methods clearly articulated? If your assessors cannot make sense of what you are proposing, your proposal has little chance of getting off the ground.
Are the proposed methods logical? In other words, do they make sense and do the assessors believe your approach can lead to credible data (generally the work of the methods section)?
Has the candidate considered the study’s boundaries as well as any potential hurdles to effective data collection and analysis? Established assessors know that all research is constrained; your job here is to acknowledge this and show the credibility of your methods in spite of any limitations (generally the work of the methods and limitations/delimitations sections).
Are the proposed methods ethical? ethics are central to all research processes (and of course the main focus of an ethics proposal). Your proposal needs to show that the dignity and well-being of respondents, both mentally and physically, are fully protected (the work of the methods and ethical considerations sections).
Are the proposed methods practical/doable? It doesn’t matter how logical and well considered your methods are if your assessors do not believe they can be implemented.
You need to show that you have or can develop the necessary expertise; that you can gain access to required data; that your timeline is realistic; and that you will come within budget (the work of the methods section as well as, if required, the timeline and budget).
Basically, your methods section needs to convince readers that your approach is an efficient, effective and ethical way to get credible answers to your questions and that you are capable of pulling this off.
What Can Other Qualitative Data Collection Methods You Consider?
So far, we have focused on the techniques and processes used in interviews, which are widely used data collection methods in qualitative research in many social science disciplines.
There are several other common qualitative data collection methods that are useful for other research purposes. Participant observation and ethnography are widely practiced data collection methods in fields of study such as anthropology and cultural geography.
These methods require careful planning and execution. The guidelines for field research are extensively covered in many discipline-specific methods blogs.
Participant Observation and Ethnography
If you intend to conduct participant observation research, there are a few things you need to consider. First, you should determine what your role is in the research field, whether it is in a community or in an organization.
The dual role of researcher-member can range from being a complete outsider researcher to being a covert participant who completely hides his/her research role from the in-group members.
Conducting research without disclosing your research purposes can entail difficult ethical dilemmas. Since you are a student, we recommend that you let your identity as a researcher be known and conduct your observation research either as an outsider researcher or as researcher-member.
This is a safer option with fewer ethical complications for student researchers.
Even when your researcher role is disclosed, ethical dilemmas are still possible, especially when you are invited to participate in group members’ routine activities that are against your own cultural norms or involve risky behaviors (e.g., binge drinking, extreme hazing, spying on someone else’s privacy).
Sometimes, you may use your status as a researcher to excuse yourself from having to participate in activities that will present an ethical dilemma for you.
But such episodes may also highlight your outsider status to the group members, which may affect what they are willing to share with you in the future.
You should understand that the line between a participant and an observer is a precarious one and be aware of the ethical and practical problems that may emerge as a result of your dual role.
Second, you will need to set aside a regular time and even space to take notes of what you have observed, reflect on them, and record your reflections. After all, you will not achieve your data collection goals if you do not set aside time for data-recording and evaluation.
You should have a notebook or an electronic device to take notes and record your observations and thoughts basically round the clock, for you never know what you may encounter or when.
Most often, you will take many short notes on the spot, as jotting down a few things is all you can do at the moment. You need to make sure that you revisit these shorthand notes (typically during a break or at the end of the day) and write them up into more detailed “field notes.”
Field notes should be detailed and including your own reflections, interpretations of things observed, and even direct quotes from conversations you had with an informant.
Third, as a researcher, you will typically develop close relationships with a few people (“key informants”) who are willing to share access to insider knowledge and help you make connections within the community. Along with leaders of the community/site you are studying you will want to conduct in-depth interviews with your key informants.
In general, you should not interfere with naturally occurring interactions within the community and follow the cultural norms and expectations in interactions with others.
Try to avoid sensitive questions or questions that will provoke group members, and you should only ask these questions of key informants with whom you have developed a trusting relationship.
In essence, focus group research is grounded on the principles of interview techniques, except that you are asking questions to a group of individuals, instead of to one interviewee.
If you plan to conduct a focus group discussion, you need to prepare an interview guide similar to the one used in interview research. Just as in an interview, you should manage focus group discussions with flexibility.
Content Analysis of Image Data
All of the data gathering techniques discussed in this section are methods for collecting text-based data. Field notes and interview transcripts produce narrative data the meaning of which will be interpreted and summarized during the data analysis phase.
With the increasing popularity of smartphone photos, YouTube videos, social media, and the abundant storage capacity of digital files, images are becoming a routine part of the social world today.
As more people use images to chronicle personal life, maintain relationships, and record historic moments, you may find images to be good data sources to address research questions about people’s attitudes, cognitive process, and experiences in contemporary society.
Images have been used as data in content analyses of magazine advertisements, films, news clips, TV shows, and other mass media. But explore other types of image data: Instagram photos, cartoons, video diaries, and drawings. You will undoubtedly find symbolic meanings embedded in images.
Your task is to interpret the social patterns and human conditions of contemporary society (e.g., an analysis of gender stereotypes in magazine advertisements, an analysis of emotional state using Instagram photos).
Or ask participants to use images to express their thoughts and attitudes (e.g., use drawings for an analysis of personality traits). While the potential is great, there are not too many models of image-based studies in the social sciences.
If you are interested in using images as your data, you will need to think creatively about how to use them to address your research questions.
The above list is a far from exhaustive list of data collection methods. Which method is most appropriate for your research depends on the objective of your study, the nature and scope of your research questions, and the feasibility of the method.
Overview and Evaluation
I add three general points here, followed by some suggestions for further readings about the qualitative approach.
1. The first point may have occurred to you if you have already had some exposure to writings about qualitative research. Those who write about the approach are certainly in general agreement concerning its basic nature and goals.
At a more specific level, however, there is often considerable divergence in the distinctions that are drawn, the points that are emphasized, and even the terminology that is used.
Adding to the diversity is the fact that the qualitative approach is by no means limited to psychology but rather figures prominently in a number of other disciplines; a partial list would include anthropology, sociology, education, marketing, and women’s studies.
Thus, the particular summary I have presented here may map only roughly onto what you encounter in other sources.
2. A second point concerns the current status of the approach in developmental psychology. Opinions on this point may of course vary. My own evaluation is that the qualitative approach is gaining in prominence and importance, but that it remains distinctly secondary to the quantitative approach in most of the field’s mainstream writing.
3. The third point concerns the fit between qualitative research and quantitative research. There are authors—including some in both the qualitative and quantitative camps—who view the approaches as incompatible and who advocate forcefully for one method of study or the other.
Most of us, however, regard the approaches as more complementary than contradictory—that is, as equally legitimate ways to address somewhat different questions about human development.
At the least, this means that our knowledge of any topic is likely to be fullest if we are aware of both the quantitative and qualitative approaches to its study.
For particular topics, it may also mean the use of both quantitative and qualitative techniques, either within the same study (the mixed methods approach discussed in the box) or within an overall program of research.
This last point about complementary approaches should sound familiar by now. Understanding human behavior is an exceedingly challenging task. It is almost always better to have multiple methods of study rather than just one.