Midterm

Question 1

Illogical Reasoning

Answer 1

Illogical reasoning occurs when we prematurely jump to conclusions or argue on the basis of invalid assumptions.

Question 2

Overgeneralization

Answer 2

Overgeneralization occurs when we unjustifiably conclude that what is true for some cases is true for all cases.

Ex. “Those people are never satisfied.”

Question 3

Selective Observation

Answer 3

Choosing to look only at things that are in line with our preferences or beliefs.

If we acknowledge only the instances that confirm our predispositions, we are victims of our own selective observation.

Ex. “Those people are never satisfied.”

Ex. Optical illusion which can be viewed as either two faces or a vase, signifying the fact that perceptions involve interpretations, and different observers may perceive the same situation differently because they interpret it differently.

Question 4

Inaccurate Observation

Answer 4

Observations based on faulty perceptions of empirical reality.

Our perceptions do not provide a direct window onto the world around us because what we think we have sensed is not necessarily what we have seen (or heard, smelled, felt, or tasted). Even when our senses are functioning fully, our minds have to interpret what we have sensed. Perceptions involve interpretations. Different observers may perceive the same situation differently because they interpret it differently.

Question 5

Resistance to Change

Answer 5

The reluctance to change our ideas in light of new information.

Resistance to change may occur for a couple of reasons: Ego-based commitments, Excessive devotion to tradition

Ex. People in many agencies who want to reject an idea use those famous words: “But we’ve never done it that way.”

Question 6

Adherence to Authority

Answer 6

Adherence to authority is given because we believe that the authority, the person making the claim, does have the knowledge. Sometimes it is difficult to change our ideas because someone in a position of authority has told us what is correct.

If we don’t have the courage to evaluate critically the ideas of those in positions of authority, we will have little basis for complaint if they exercise their authority over us in ways we do not like. And if we do not allow new discoveries to call our beliefs into question, our understanding of the social world will remain limited.

Ex. “You’re wrong about the impact of structural issues on economic well-being. My parents said that anyone can get ahead in life if they want to.”

Question 7

The Social Scientific Approach

Answer 7

Social science is the use of scientific methods to investigate individuals, groups, communities, organizations, societies, and social processes; the knowledge produced by these investigations.

The scientific approach to answering questions about the social world is designed to greatly reduce these potential sources of error in everyday reasoning. Social science relies on logical and systematic methods to answer questions, and it does so in a way that allows others to inspect and evaluate its methods.

Scientific research develops a body of knowledge that is continually refined as beliefs are rejected or confirmed on the basis of testing empirical science.

Social work research relies on these methods to investigate treatment effectiveness, social conditions, organizational behavior, and social welfare policy.

Social science methods can reduce the risk of selective, inaccurate, or incomplete observation by requiring that we measure and sample phenomena systematically.

Question 8

Social Work Research in Practice

Answer 8

There are a great many studies of different phenomena, social conditions, impacts of different programs, and intervention methods, we can classify the purposes of these studies into four categories: description, exploration, explanation, and evaluation.

Question 9

Descriptive Research

Answer 9

Research in which social phenomena are defined and described.

Descriptive research typically involves the gathering of facts. Measurement and sampling are central concerns in descriptive research. Survey research is often used for descriptive purposes.

Question 10

Exploratory Research

Answer 10

Exploratory research seeks to find out how people get along in the setting under question, what meanings they give to their actions, and what issues concern them.

The goal is to figure out “what is going on here” and to investigate social phenomena without explicit expectations. The purpose is associated with the use of methods that capture large amounts of relatively unstructured information. Research like this frequently involves qualitative methods.

Ex. How do the homeless adapt to shelter life?

Question 11

Explanatory Research

Answer 11

Explanatory research seeks to identify causes and effects of social phenomena and to predict how one phenomenon will change or vary in response to variation in some other phenomenon.

Explanatory research depends on our ability to rule out other explanations for our findings, to demonstrate a time order between two events, and to show that the two events are related to each other. Research methods used to identify causes are effects are the focus of Ch. 6.

Ex. What community-level factors cause homelessness?

Question 12

Evaluation Research

Answer 12

Evaluation research is research that describes or identifies the impact of social programs and policies.

Ex. Should housing or treatment come first?

Evaluation research involves searching for practical knowledge in considering the implementation and effects of social policies and the impact of programs.

Question 13

Alternative Research Strategies

Answer 13

When conducting social work research, we are attempting to connect theory with empirical data — the evidence we obtain from the social world. Researchers use two alternative strategies to make this connection: 1) Deductive research and 2) Inductive research.

Question 14

Deductive Research

Answer 14

Deductive research starts with a theory and then some of its implications are tested with data; it is most often the strategy used in quantitative methods.

Starting with a theory, a specific expectations is derived, data are collected to test the specific expectation.

Ex. When people have more human capital (work-related skills and education) they are likely to have higher incomes. From this relationship we can deduce a hypothesis, or a more specific expectation, that person who graduate from college should have a higher incomes than persons who do not graduate from college. Now that we have a hypothesis, we can collect data about level of education and income. We can’t always directly test the general theory but we can test specific hypotheses that are deducted from it.

Question 15

Hypothesis

Answer 15

A tentative statement about empirical reality, involving a relationship between two or more variables.
Ex. The higher the poverty rate in a community, the higher the percentage of community residents who are homeless.

Variation in one variable is proposed to predict or cause variation in the other variable. Ex. Having a college degree or having less than a college degree will predict income level. College graduate is the proposed influence called the independent variable; its effect or consequence, in this case income level, is the dependent variable.

A hypothesis derived from a theory doesn’t just state that there is a connection between variables, it suggests that one variable actually influences another - that a change in the first one somehow predicts, influences, or causes a change in the second. It says that if one thing happens, then another thing is likely to happen.

Question 16

Variables

Answer 16

Characteristics or property that can take on different values or attributes.
Ex. Poverty rate, percentage of homeless community residents.

Question 17

Independent Variable

Answer 17

A variable that is hypothesized to cause, or lead to, variation in another variable.
Ex. Poverty rate

Question 18

Dependent variable

Answer 18

A variable that is hypothesized to vary depending on or under the influence of another variable.
Ex. Percentage of community residents who are homeless.

Question 19

Direction of Association

Answer 19

A pattern in a relationship between two variables; the values of one variable tend to change consistently in relation to change in the value of the second variable.

Ex. An increase in the independent variable might lead to an increase in the dependent variable or an increase in the independent variable might predict a decrease in the dependent variable. When one variable increases as the other increase, the direction of association is positive, when its vice covers a, the direction of association is negative (or inverse).

Question 20

Measurement Validity

Answer 20

Measurement validity is our first concern in establishing the validity of research results because if we have not measured what we think we measured, we really do not know what we are talking about.

Measurement validity exists when a measure measures what we think it measures.

Ex. Must be careful to ensure that the measures are comparable for diverse groups, we cannot just assume the measures are valid for all subgroups of a population.

Question 21

Generalizability

Answer 21

The generalizability of a study is the extent to which it can be used to inform us about people, places, or events that were not studied.

Question 22

Sample Generalizability/Cross-population generalizability

Answer 22

The two kinds of generalizability - sample generalizability refers to the ability to take findings obtained from a sample or a subset of a larger population and apply them to that population. Ex. A community organizer may study a sample of residents living in a particular neighborhood in order to assess their attitudes toward opening a homeless shelter in their neighborhood then generalize the findings to all the residents of the neighborhood.

Cross-population generalizability refers to the ability to generalize from findings about one group or population or setting to other groups or populations or settings. Cross-population generalizability exists when findings about one group or population or setting hold true for other groups or populations or settings.

Ex. If we pull a representative sample from a population, we can generalize the sample results to the population from which the sample was selected, but we should be cautious in generalizing to another setting or population.

Question 23

From Concepts to Observations

Answer 23

After we define the concepts in a study, we must identify corresponding variables and develop procedures to measure them. Ex. To measure alcohol abuse we might use any number of variables: one variable might be the count of alcoholic drinks; another might involve asking about the presence of blackouts; a third variable may ask about binge drinking; and a fourth variable might reflect a score on a rating scale of 10 questions. All of these variable could show low or high degrees of substance abuse.

Question 24

Operationalization

Answer 24

The process of specifying the operations that will indicate the value of cases on a variable.

Once we have defined our concepts in the abstact — that is, we have provided a nominal definition— and we have identified the specific variables we want to measure, we must develop measurement procedures. The goal is to devise procedures to indicate the values of cases on a variable. Operationalization is process of connecting concepts to observations. An operational definition provided by researchers includes what is measured, how the indicators are measured, and the rules used to assign a value to what is observed and to interpret the value.

Question 25

Concepts

Answer 25

A mental image that summarizes a set of similar observations, feelings, or ideas. Ex. Binge drinking, mental health, poverty, alcohol abuse.

Concepts must be defined to make them useful in research. There should be only one definition of a concept that we have to specific clearly what we mean when we use a concept.

Question 26

Levels of Measurement

Answer 26

The final part of Operationalization is to assign a value or symbol to represent the observation. Each variable has categories of some sort and we need to know how to assign a symbol (typically a number) to represent what has been observed or learned.

When we know a variable’s level of measurement, we can better understand how cases vary on that variable and so understand more fully what we have measured.

There are 4 levels of measurement: nominal, ordinal, interval, and ratio.

Levels of measurement has important implications for the type of mathematical procedures and statistics that can be used with the variable.

Question 27

Nominal Level of Measurement

Answer 27

The nominal level of measurement identifies variables whose values have no mathematical interpretation. They vary in kind or quality, but not in amount.

They may also be called categorical variables. They’re often referred to as attributes instead of values. Ex. Ethnicity as a variable can have several attributes (or categories or qualities): African American, Hispanic, Asian American, White, Native American… etc.

Nominal-level variables are commonplace in SW research. Client characteristics such as marital status, or mental health diagnosis, are commonplace nominal-level variables. They are qualitative variables. Their order has no meaning mathematically.

The only mathematical operation we can perform with nominal-level variables is a count. Ex. We can count how many clients last month were females and how many were males.

Attributes of categorical variables must be assigned to cases with great care. The attributes we use to categorize cases must be mutually exclusive and exhaustive. A variables attributes or values are mutually exclusive if every case can have only one attribute. A variables attributes or values are exhaustive when every case can be classified into one of the categories.

Changes in social conventions can make appropriate classification at the nominal level much more complicated. Ex. “Race” and measuring it is not so straightforward.

Question 28

Ordinal Level of Measurement

Answer 28

The first of the three quantitative levels is the ordinal level of measurement. It is a measurement of a variable in which the numbers indicating a variable’s values specify only the order of the cases, permitting greater than and less than distinctions.

The different values measured at the ordinal level must be mutually exclusive and exhaustive and must cover the range of observed values and allow each case to be assigned no more than one value.
Ex. Cup sizes at a coffee show - small, medium, or large. The categories represent relative cup sizes but the gaps between the various responses don’t have any particular meaning whether in quantity or in price.

Ex. Goal attainment scales to measure the progress of a client in achieving a particular goal. Scales are usually developed by describing the worst indicators, the best indicators, and several steps in between with the gaps having no meaning and the scoring representing the progress of the client. There is an order to the levels of achievement and we can describe how many clients fall into each category.

Question 29

Interval Level of Measurement

Answer 29

The values of a variable at the interval level of measurement represent fixed measurement units but have no absolute or fixed zero point.

Also has mutually exclusive categories, the categories are exhaustive, and there is an order to the responses. The gaps between the numbers of the scale are meaningful; a one-unit difference is the same at any point in the scale. Ex. The difference between two Fahrenheit temperatures. Because the gaps between the numbers are equal on this scale, the gap between 60 degrees and 30 degrees is actually 30, but 60 in this vase is not twice as hot as 30. Because heat does not begin at 0 degrees on the Fahrenheit scale.

The zero value on an interval scale does not indicate the complete absence of the measured variable.

There are more mathematical operations associated with interval-level variables.

Question 30

Ratio Level of Measurement

Answer 30

A measurement of a variable in which the numbers indicating a variable’s values represent fixed measuring unit and an absolute zero point.

In this case, zero means absolutely no amount of whatever the variable indicates.

Ex. The number of clients in a program, the time spent providing counseling, or the number of hot meals delivered to homebound elderly. In each case the answer zero is meaningful, representing the complete
absence of the variable.

Ratio numbers can be added and subtracted because the numbers begin at an absolute zero point and they can also be multiplied and divided (so that ratios can be formed between the numbers. Ex. People’s ages can be represented by values ranging from 0 years (or some fraction of a year) to 120 or more. A person who is 30 years old is 15 years older than someone who is 15 years old and is twice as old as that person. Of course the numbers also are mutually exclusive, are exhaustive, have an order, and there are equal gaps.

Question 31

The Case of Dichotomies

Answer 31

Dichotomies, variables having only 2 values, are a special case from the standpoint of levels of measurement. The values or attributes of a variable such as depression clearly vary in kind or quality, not in amount. Thus the variable is categorical — measured at the nominal level.

Yet in practical terms, we can think of the variable in a slightly different way, as indicating the presence of the attribute depressed or not depressed. Viewed in this way, there is an inherent order: A depressed person has more of the attribute (it is present) than a person who is not depressed (the attribute is not present). We are likely to act given the presence or absence of that attribute; we intervene or refer to treatment a depressed client, whereas we would not do so with a client who was not depressed. Nonetheless, although in practical terms there is an order empirically we treat a dichotomous variable as a nominal variable.

Question 32

Assessment of Measurement Accuracy

Answer 32

Do the operations to measure our variables provide stable or consistent responses — are they reliable?

Do the operations developed to measure our concepts actually do so — are they valid?

When we test the effectiveness of two different interventions or when we monitor a client’s progress, we want the changes we observe to be due to the intervention and not to the instability or inaccuracy of the measurement instrument. We also want to know that the measure we use is really a measure of the outcome and not a measure of some other outcome. We cannot have much confidence in a measure until we have empirically evaluated its reliability and validity.

Question 33

Reliability

Answer 33

Reliability means that a measurement procedure yields consistent or equivalent scores when the phenomenon being measured is not changing. If a measure is reliable, it is affected less by random error or chance variation than if it is unreliable.

Reliability is a prerequisite for measurement validity.

Test-Retest Reliability - if you take a test of your research methodology knowledge and retake the test 2 months later, the test is performing reliably if you receive a similar score both times — assuming nothing changed in your research methodology knowledge during those 2 months.

Question 34

Validity

Answer 34

Validity refers to the extent to which measures indicate what they are intended to measure. Technically, a valid measure of a concept is one that is (a) closely related to other apparently valid measures of the concept, (b) closely related to the known or supposed correlates of that concept, and (c) not related to measures of unrelated concepts (adapted from Brewer & Hunter, 2005). Measurement validity is assessed with four different approaches: face validation, content validation, criterion validation, and construct validation.

Face validity The type of validity that exists when an inspection of items used to measure a concept suggests that they are appropriate “on their face.”For example, a count of how many drinks people consumed in the past week would be a face-valid measure of their alcohol consumption.

Content validity The type of validity that exists when the full range of a concept’s meaning is covered by the measure.

Criterion validity The type of validity established by comparing the scores obtained on the measure being validated to scores obtained with a more direct or already validated measure of the same phenomenon (the criterion).
Concurrent validity The type of validity that exists when scores on a measure are closely related to scores on a criterion measured at the same time.
Predictive validity The type of validity that exists when a measure predicts scores on a criterion measured in the future.

Question 35

Sampling Components and the Population

Answer 35

We often do not have the time or resources to study the entire population, that is, the entire set of individuals (or other entities) in which we are interested. Therefore, we decide to study a sample, a subset of the population of interest. The individual members or other entities of the sample are called elements.In many studies, we sample directly from the elements in the population of interest. We may survey a sample of the entire population of students at a school based on a list obtained from the registrar’s office. This list from which the elements of the population are selected is termed the sampling frame.

Question 36

Population

Answer 36

The entire set of individuals or other entities to which study findings are to be generalized.

Question 37

Sample

Answer 37

A subset of a population that is used to study the population as a whole.

Question 38

Elements

Answer 38

The individual members of the population whose characteristics are to be measured.

Question 39

Sampling Frame

Answer 39

A list of all elements or other units containing the elements in a population.

Question 40

Enumeration Units

Answer 40

Units that contain one or more elements and that are to be listed in a sampling frame.

Question 41

Sampling Units

Answer 41

Units listed at each stage of a multistage sampling design.

Ex. Sample of child welfare agencies (Agencies are the elements and the primary sampling unit) - sample of social work staff in the agencies (Social work staff are the secondary sampling units; they provide information about the agencies).

Question 42

Sampling Methods

Answer 42

Certain features of samples make them more or less likely to represent the population from which they are selected. The most important distinction to be made is whether the samples are based on a probability or nonprobability sampling method. Probability sampling methods allow us to know in advance how likely it is that any element of a population will be selected for the sample. Sampling methods that do not let us know the likelihood in advance are termed nonprobability sampling methods.

Question 43

Probability Sampling

Answer 43

Probability sampling methods are sampling methods that rely on a random or chance selection method so that the probability of selection of population elements is known.

Probability sampling methods are those in which the probability of selection is known and is not zero, so there is some chance of selecting each element. These methods randomly select elements and therefore have no systematic bias; nothing but chance determines which elements are included in the sample. This feature of probability samples makes them much more desirable than nonprobability samples when the goal is to generalize to a larger population.

The larger the sample, the more confidence we can have in the sample’s representativeness: If we randomly pick five people to represent the entire population of our city, our sample is unlikely to be representative of the entire population in terms of age, gender, race, attitudes, and so on. But if we randomly pick 100 people, the odds of having a representative sample are much better; with a random sample of 1,000, the odds become very good indeed.

The more homogeneous the population, the more confidence we can have in the representativeness of a sample of any particular size. Let’s say we plan to draw samples of 50 from each of two communities to estimate mean family income. One community is diverse, with family incomes varying from $12,000 to $85,000. In the other, more homogeneous community, family incomes are concentrated in a narrow range, from $41,000 to $64,000. The estimated mean family income based on the sample from the homogeneous community is more likely to be representative than is the estimate based on the sample from the more heterogeneous community. With less variation to represent, fewer cases are needed to represent the homogeneous community.

The fraction of the total population that a sample contains does not affect the sample’s representativeness unless that fraction is large. The number of cases is more important than the proportion of the population represented by the sample. We can regard any sampling fraction less than 2% with about the same degree of confidence (Sudman, 1976). In fact, sample representativeness is not likely to increase much until the sampling fraction is quite a bit higher. Other things being equal, a sample of 1,000 from a population of 1 million (with a sampling fraction of 0.001, or 0.1%) is much better than a sample of 100 from a population of 10,000 (although the sampling fraction is 0.01, or 1%, which is 10 times higher). The size of the sample is what makes representativeness more likely, not the proportion of the whole that the sample represents.

Question 44

Simple Random Sampling

Answer 44

Simple random sampling requires some procedure that generates numbers or otherwise identifies cases strictly on the basis of chance. As you know, flipping a coin and rolling a die both can be used to identify cases strictly on the basis of chance, but these procedures are not efficient tools for drawing a sample. A random numbers table, like the one in Exhibit 5.4 simplifies the process considerably. The researcher numbers all the elements in the sampling frame and then uses a systematic procedure for picking corresponding numbers from the random numbers table. (Practice Exercise 1 at the end of this chapter explains the process step by step.) Alternatively, a researcher may use a lottery procedure. Each case number is written on a small card, and then the cards are mixed up and the sample selected from the cards.

Simple random sampling is a sampling method in which every sample element is selected only on the basis of chance through a random process.

When a large sample must be generated, these procedures are cumbersome. For a large sample, a computer program can easily produce a random sample of any size by generating a random selection of numbers within the desired range. Random number generators may also be found on the Internet simply by searching using random numbers generator.

Question 45

Systematic Random Sampling

Answer 45

A sampling method in which sample elements are selected from a list or sequential files, with every nth element being selected after the first element is selected randomly with the first interval.

Systematic random sampling is a variant of simple random sampling. The first element is selected randomly from a list or from sequential files and then every nth element is selected. This is a convenient method for drawing a random sample when the population elements are arranged sequentially. It is particularly efficient when the elements are not actually printed (i.e., there is no sampling frame), but instead are represented by folders in filing cabinets.
Systematic random sampling requires three steps:
1. The total number of cases in the population is divided by the number of cases required for the sample. This division yields the sampling interval, the number of cases between one sampled case and another. If 50 cases are to be selected out of 1,000, the sampling interval is 20; every 20th case is selected.
2. A number from 1 to 20 (or whatever the sampling interval is) is selected randomly. This number identifies the first case to be sampled, counting from the first case on the list or in the files. Alternatively, a number is selected randomly using the entire range; in this case, from 1 to 1,000. In either method, a random numbers table or a random number generator can be used to decide on a starting number.
3. After the first case is selected, every nth case is selected for the sample, where n is the sampling interval. If the sampling interval is not a whole number, you may round, but whatever the decimal is, you must round up even if the interval is 30.1. Rounding down precludes some elements from having any chance of being selected. Alternatively you may vary the size of the sampling interval to yield the proper number of cases for the sample. For example, if the sampling interval is 30.5, the sampling interval alternates between 30 and 31.

Question 46

Sampling Interval

Answer 46

The number of cases from sampled case to another in a systematic random sample.

Question 47

Stratified Random Sampling

Answer 47

Stratified random sampling uses information known about the total population prior to sampling to make the sampling process more efficient.

First, all elements in the population (i.e., in the sampling frame) are distinguished according to their value on some relevant characteristic. This characteristic forms the sampling strata. For example, race may be the basis for distinguishing individuals in some population of interest. Next, elements are sampled randomly from within these strata; so within each racial category, individuals are randomly sampled. Of course, using this method requires more information prior to sampling than is the case with simple random sampling. Each element must belong to one and only one stratum and the size of each stratum in the population must be known.This method is more efficient than drawing a simple random sample because it ensures appropriate representation of elements across strata.

Imagine that you plan to draw a sample of 500 from an ethnically diverse neighborhood. The neighborhood population is 15% Black, 10% Hispanic, 5% Asian, and 70% White. If you drew a simple random sample, you might end up with disproportionate numbers of each group. But if you created sampling strata based on race and ethnicity, you could randomly select cases from each stratum: 75 Blacks (15% of the sample), 50 Hispanics (10%), 25 Asians (5%), and 350 Whites (70%). By using proportionate stratified sampling, you would eliminate any possibility of error in the sample’s distribution of ethnicity. Each stratum would be represented exactly in proportion to its size in the population from which the sample was drawn

Question 48

Proportionate Stratified Sampling

Answer 48

Sampling method in which elements are selected from strata in exact proportion to their representation in the population.

By using proportionate stratified sampling, you would eliminate any possibility of error in the sample’s distribution of ethnicity. Each stratum would be represented exactly in proportion to its size in the population from which the sample was drawn

Question 49

Disproportionate Stratified Sampling

Answer 49

Sampling in which elements are selected from strata in different proportions from those that appear in the population.

In disproportionate stratified sampling, the proportion of each stratum that is included in the sample is intentionally varied from what it is in the population. In the case of the sample stratified by ethnicity, you might select equal numbers of cases from each racial or ethnic group: 125 Blacks (25% of the sample), 125 Hispanics (25%), 125 Asians (25%), and 125 Whites (25%). In this type of sample, the probability of selection of every case is known but unequal between strata. You know what the proportions are in the population, and so you can easily adjust your combined sample statistics to reflect these true proportions. For instance, if you want to combine the ethnic groups and estimate the average income of the total population, you would have to weight each case in the sample. The weight is a number you multiply by the value of each case based on the stratum it is in. For example, you would multiply the incomes of all Blacks in the sample by 0.6 (75/125), the incomes of all Hispanics by 0.4 (50/125), and so on. Weighting in this way reduces the influence of the oversampled strata and increases the influence of the undersampled strata to just what they would have been if pure probability sampling had been used.

Question 50

Cluster Sampling

Answer 50

Cluster sampling is useful when a sampling frame is not available, as is often the case for large populations spread out across a wide geographic area or among many different organizations. A cluster is a naturally occurring mixed aggregate of elements of the population, with each element appearing in one and only one cluster. Schools could serve as clusters for sampling students, blocks could serve as clusters for sampling city residents, counties could serve as clusters for sampling the general population, and agencies could serve as clusters for sampling social work staff.

Cluster sampling is at least a two-stage procedure. First, the researcher draws a random sample of clusters. A list of clusters should be much easier to obtain than a list of all the individuals in each cluster in the population. Next, the researcher draws a random sample of elements within each selected cluster. Because only a fraction of the total clusters is involved, obtaining the sampling frame at this stage should be much easier.

For example, in a needs assessment of residents of a particular neighborhood, blocks could be the first-stage clusters. Someone could walk around each selected block and record the addresses of all occupied dwelling units (see Exhibit 5.7). In a cluster sample of students, a researcher could contact the schools selected in the first stage and make arrangements with the registrar to obtain lists of students at each school. Cluster samples often involve multiple stages, with clusters within clusters, as when a national sample of middle school students involves first sampling states, then counties, then schools, and finally students in each selected school.

Question 51

Non probability Sampling Methods

Answer 51

Nonprobability sampling methods do not use a random selection procedure, and therefore, elements within the population do not have a known probability of being selected. We cannot expect a sample selected using a nonprobability sampling method to yield a representative sample. The findings cannot be generalized to the broader population of interest.

Nonprobability sampling methods are often used in qualitative research. In qualitative research, a focus on one setting or a very small sample allows a more intensive portrait of activities and actors. In quantitative research, these methods are useful when random sampling is not possible, with a research question that does not concern a large population or require a random sample or for a preliminary pilot study. These methods are often applied to experimental studies testing the effectiveness of different treatment or intervention methods or with program evaluations conducted in agencies. There are four commonly used nonprobability sampling methods: (1) availability sampling, (2) quota sampling, (3) purposive sampling, and (4) snowball sampling. Because nonprobability sampling methods do not use a random selection procedure, we cannot expect a sample selected with any of these methods to yield a representative sample.

Question 52

Availability (Convenience) Sampling

Answer 52

Elements are selected for availability sampling (or convenience sampling) because they are available or easy to find. There are many ways to select elements for an availability sample: standing on street corners and talking to whoever walks by, asking questions of employees who come to pick up their paychecks at a personnel office and who have time to talk to a researcher, or approaching particular individuals while observing activities in a social setting. To study sexual risk taking among homeless youth in Minneapolis, Linda Halcón and Alan Lifson (2004) hired experienced street youth outreach workers who approached youth known or suspected to be homeless and asked whether they would be willing to take part in an interview. To describe why women become homeless and how these reasons might differ from male homelessness, Tara Richards and her colleagues (T. N. Richards, Garland, Bumphus, & Thompson, 2010) went to homeless shelters, soup kitchens, and a shelter to find participants.

Question 53

Quota Sampling

Answer 53

Quota sampling is intended to overcome the most obvious flaw of availability sampling—that the sample will consist of only whoever or whatever is available, without any concern for its similarity to the population of interest. The distinguishing feature of quota sampling is that quotas are set to ensure that the sample represents certain characteristics in proportion to their prevalence in the population.Suppose that you wish to sample adult residents of a city in a study of support for building a casino. You know from the city’s annual report what the proportions of the residents are in terms of gender, employment status, age, and race. You think that each of these characteristics might influence support for building a casino, so you want to be sure that the sample includes men, women, people who work, people not in the labor force, older people, younger people, and various ethnic groups in proportion to their numbers in the town population.

Question 54

Purposive Sampling

Answer 54

In purposive sampling, each sample element is selected for a purpose usually because of the unique position of the sample elements. Purposive sampling may involve studying the entire population of some limited group (directors of shelters for homeless adults) or a subset of a population (mid-level managers with a reputation for efficiency). Purposive sampling may be used to examine the effectiveness of some intervention with a set of subjects or clients who have particular characteristics, such as a specific diagnosis. A purposive sample may be a key informant survey, which targets individuals who are particularly knowledgeable about the issues under investigation.

Question 55

Snowball Sampling

Answer 55

For snowball sampling, you identify one member of the population and speak to him or her; you ask that person to identify others in the population and speak to them; you ask them to identify others, and so on. The sample thus “snowballs” in size. Snowball sampling is useful for hard-to-reach or hard-to-identify, interconnected populations (at least some members of the population know each other), such as drug users, parents with small children, participants in Alcoholics Anonymous groups or other peer support groups, and informal organizational leaders. For example, Suk-Young Kang and his colleagues (Kang, Domanski, & Moon, 2009) used snowball sampling to learn about depression among Korean older adult immigrants who did not reside in ethnic enclaves. Because there was no sampling frame, they sought participants from a Korean church and then asked those whom they had recruited to refer other older adult Koreans to the study.

However, researchers using snowball sampling normally cannot be confident that their sample represents the total population of interest.One caveat when using a snowball sampling technique is that you are asking people to identify other people with a similar status without the knowledge or consent of the people being identified. The people who are identified may not wish others to know that they have a particular status. This is particularly a concern when snowball sampling is used to identify subgroups of the population who may experience oppression or discrimination because they hold a particular status. In class, we often use a sampling exercise that requires students to identify a nonprobability sampling technique to gather information from gay and lesbian members of the community with the purpose of identifying their social service needs. Often students will suggest snowball sampling without realizing that what they are doing is asking people to “out” their acquaintances without permission of those being identified.