LoBiondo-Wood - Nursing Research in Canada

Chapter 12: Sampling LoBiondo-Wood: Nursing Research in Canada MULTIPLE CHOICE 1. Which of the following is the process of selecting representative units of a population for a research investigation? a. Sampling b. Snowballing c. Delimitation d. Random assignment ANS: A A B C D Feedback Sampling is the process of selecting representative units of a population for a research study. Snowballing is a sampling technique that uses social networks. Delimitation is the specification of sample characteristics. Random assignment is the process of assigning participants to groups. 2. How should a nurse researcher expect a sample to differ from a population? a. A sample can be an object or an event; a population refers to individuals or groups of people. A population has a broad set of defining characteristics, whereas a sample has a narrow set of defining characteristics. c. A population is a representative segment of a defined sample. d. A sample is a representative segment of a defined population. b. ANS: D A B C D Feedback Population refers to a well-defined set of people, animals, objects, or events. A sample is a group of research participants representing a population. A sample is a representative segment of a population. A sample is a set of elements or segments of a specified population. 3. A nurse researcher has made a generalization on the basis of the experience of a small number of participants. What will the result of this be? a. Small samples invalidate hypotheses. b. The researcher may be unable to eliminate his or her bias. c. Data obtained from a small number may represent the phenomenon inadequately. d. Small samples increase historical threats to internal validity. ANS: C A B C D Feedback A small sample does not necessarily invalidate hypotheses. Researcher bias is not necessarily controlled by sample size, small or big. A very small sample may represent the phenomenon inadequately and threaten valid conclusions. A small sample size does not necessarily increase historical threats to validity. 4. Which of the following statements best identifies the sample descriptors that form the basis for eligibility criteria in sample selection? a. A sample of 1,000 practising nurses in a selected state was surveyed. b. The sample consisted of 267 Canadian girls and American girls who lived in urban areas in their respective countries. c. The sample consisted of 40 men between the ages of 20 and 55 who were treated for burn injuries during the past decade at a specialized burn unit. d. The sample consisted of 57 hospital nurse executives and 178 staff registered nurses employed in Ontario. ANS: C A B C D Feedback The description includes only practising nurses and a selected state. The description includes only gender, ethnicity, and urban dweller setting. Sample descriptors should explicitly describe the population criteria for inclusion (eligibility); this sample descriptor has four specified criteria. The description includes nurse executives or nurses from one province. 5. Which of the following statements about inclusion and exclusion criteria is true? a. These criteria are not needed when the sample being studied is composed of events. These criteria serve to ensure that the characteristics of the sample and the population are congruent. c. Inclusion criteria are used to define independent variables, and exclusion criteria are used to define dependent variables. d. Inclusion criteria are used to define dependent variables, and exclusion criteria are used to define independent variables. b. ANS: B A B C D Feedback Inclusion and exclusion criteria are always needed in some form. Inclusion and exclusion criteria are designed to keep both population and sample characteristics congruent. Inclusion and exclusion criteria refer to sample participants, rather than to independent or dependent variables. Inclusion and exclusion criteria refer to sample participants, rather than to independent or dependent variables. 6. What can a nurse researcher use to restrict the study population to homogeneous groups of participants? a. Sampling b. Self-selection c. Eligibility criteria d. Defining study limitations ANS: C A B C D Feedback Sampling does not mandate restricting the population to a homogeneous group of participants. Self-selection would not restrict participants to a homogeneous group. Eligibility criteria restrict or control the homogeneity of participants in relation to the population. Specifying study limitations would not restrict participants to homogeneity. 7. A nurse researcher should be concerned about establishing clear eligibility criteria for inclusion in a study sample because such criteria will do what? a. Increase the homogeneity of the sample b. Decrease the homogeneity of the sample c. Increase the size of the sample d. Decrease the size of the sample ANS: A A B C D Feedback Clear eligibility criteria are designed to increase the homogeneity of the sample. Clear eligibility would not hinder homogeneity but support it. Eligibility criteria do not directly increase or decrease sample size. Eligibility criteria do not directly increase or decrease sample size. 8. A nurse researcher should attempt to clearly specify exclusion criteria for a study sample because such criteria will do what? a. Limit the time needed to complete the study b. Reduce the effect of extraneous variables on the accurate evaluation of the outcome variable Eliminate the outside influences and inferences regarding the effectiveness of a behavioural intervention d. Ensure that an adequate sample size is accrued to meet assumptions for accurate statistical analysis c. ANS: B A B C D Feedback Exclusion criteria do not directly affect how long a study takes. Exclusion criteria are designed to control sample characteristics and thereby reduce the potential effect of extraneous variables on the outcome variable. Exclusion criteria delimit the sample but do not fully act as controls for the intervention. Sample size is not determined by exclusion criteria. 9. The nurse researcher knows that a sample is representative of a population when which statement is true? a. The sample is homogeneous. b. All units of a population are included. c. The most readily accessible persons are used as participants. d. The characteristics of the sample closely approximate those of the population. ANS: D A B C D Feedback A homogeneous sample is not the ultimate criterion for evaluating a sample. Population elements or unit inclusion is important but not the ultimate criterion. Ready participant accessibility does not constitute the most important evaluative criteria. The foremost criterion in evaluating a sample is its representativeness. 10. What is the difference between an “accessible population” and a “target population”? a. An accessible population meets inclusion criteria, whereas a target population meets exclusion criteria. A target population meets inclusion criteria, whereas an accessible population meets exclusion criteria. c. An accessible population represents the entire set of cases the researcher wishes to study, whereas a target population represents that part of the accessible population that could feasibly be included in the study. d. A target population represents the entire set of cases the researcher wishes to study, whereas an accessible population represents that part of the target population that could feasibly be included in the study. b. ANS: D A B C D Feedback Inclusion and exclusion criteria refer to participant characteristics, not to components of a population. Inclusion and exclusion criteria refer to desirable and undesirable participant characteristics. A target population is the entire set of cases to be studied; an accessible population is a part of the population that can be studied. A target population is the entire set of cases the researcher wants to study; an accessible population meets population criteria and is available for study. 11. Which factor or condition of a sample meets the criterion for representativeness? a. The sample is drawn randomly from the general population without the use of specific criteria. The characteristics of the sample closely resemble the characteristics of the larger target population. c. The sample includes the same percentage of men and women as well as members of different ethnic groups as the general population does. d. The participants selected for the sample have agreed to express the specific opinions or beliefs representative of their ethnic or cultural group. b. ANS: B A B C D Feedback Representativeness is based directly on the use of specific criteria. A sample is considered representative when its characteristics closely resemble those of the general population. These factors are not as critical as is closely matching the characteristics of the general population. Participants’ agreement with the attitudes or beliefs of their ethnic or cultural groups is not considered representativeness. 12. What is the major benefit of nonprobability sampling, as compared with probability sampling? a. It does not affect generalizability. b. Informed consent must be obtained. c. The necessary sample sizes are easier to obtain. d. Sample sizes are too small for most methods of statistical analysis. ANS: C A B C D Feedback Nonprobability sampling reduces or limits generalizability. Informed consent is required for all studies despite the choice of sample. Convenience nonprobability sampling is more accessible. Nonprobability sampling does not mandate a small sample size. 13. Which of the following sampling strategies has the greatest risk for sample bias? a. Quota b. Random c. Purposive d. Convenience ANS: D A B C D Feedback Quota sampling, if done appropriately, will not create sample bias. Random sampling should help avoid sample bias. Purposive sampling would enable selective inclusion of research participants and help avoid sampling bias. Convenience sampling has the greatest risk for sample bias. 14. At a teaching hospital, of the staff of 1,500 nurses, 60% have a college diploma, and 21% have a BScN. The recruitment goal for a study was set at 1,000 nurses, with 600 in the college diploma group, 210 in the BScN group, and 190 from any other type of nursing education. What type of sampling strategy is represented by this research plan? a. Quota b. Random c. Purposive d. Convenience ANS: A A B C D Feedback In quota sampling, the researcher ensures that proportional segments of the population are included in the sample. Random sampling would not ensure that specified proportions of the population were included in the sample. Purposive sampling ensures inclusion of participants’ characteristics but not proportional segments of the population. Convenience sampling includes easily available research participants, not proportional segments of the population. 15. A study was conducted to determine whether head flattening in preterm infants was prevented by placing their heads on a water pillow. The first 50 preterm infants admitted to a newborn intensive care unit during a specific year were assigned to either a usual care group or a treatment group. What type of sampling strategy is represented by this research plan? a. Quota b. Random c. Purposive d. Stratified ANS: C A B C D Feedback Quota sampling ensures proportional segments in the sample that mirror the population, but it does not include random sampling. Random sampling ensures that all participants in a population have an equal chance of selection. In purposive sampling, the researcher hand picks cases based on his or her knowledge of the population (preterm infants). Stratified sampling is similar to quota sampling and includes random sampling. 16. What is the primary characteristic of a probability sample? a. Self-selection of participants b. Random selection of the sample c. Hand picking of participants by the researcher d. Representation of proportional segments of the population ANS: B A B C D Feedback Self-selection of participants would be antithetical to probability sampling. The primary characteristic of a probability sample is random selection. Participants hand picked by a researcher represent purposive sampling, which is not a form of probability sampling. Proportional segment representation is found in quota sampling, which is a form of nonprobability sampling. 17. What is the major characteristic of a stratified sample? a. The sample population is divided into subsets that are homogeneous for a particular trait or feature. The sample population is divided into subsets for random assignment into an intervention group or a usual care group. c. The sample population is randomly assigned to a specific subgroup that will differ from other groups on the basis of treatment type. d. The sample population is randomly assigned to a specific subgroup that will differ from other groups on the basis of treatment duration. b. ANS: A A B C D Feedback The major characteristic of a stratified sample is that selected participants represent population subgroups that are homogeneous. A stratified sample would have representation in both the treatment group and the control group. In a stratified sample, both the treatment group and the control group are representative of population subgroups. Stratified samples are not devised on the basis of length of treatment but on the proportions of participants in the population. 18. What type of sampling strategy is represented when every fourth postsecondary student who registers for a specific course is selected into the treatment group? a. Cluster sampling b. Systematic sampling c. Simple random sampling d. Stratified random sampling ANS: B A B C D Feedback Cluster sampling involves sampling selected segments or settings of a population (e.g., all urban hospitals). Systematic sampling involves selection of every Kth case (in this instance every fourth). Simple random sampling involves all members of the population having an equal chance of selection, not selection by placement. Stratified random sampling involves selecting proportional segments of the sample representing the population through random selection. 19. What is the appropriate sampling interval for drawing a systematic sample of 25 participants from 200 women who had breast enhancement surgery in the course of a particular year at a specific medical centre? a. Every fourth patient b. Every fifth patient c. Every eighth patient d. Every tenth patient ANS: C A B C D Feedback The sampling interval is determined by dividing the total sample by a standard distance between elements. The sampling interval is determined by dividing the total sample by a selected standard distance between elements. The sampling interval would be determined by dividing 200 by 25; that is, every eighth patient. The sampling interval is determined by dividing the total sample by a selected standard distance between elements. 20. A nurse researcher would hesitate to use nonprobability sampling because a. it will result in reduced or limited generalizability. b. informed consent must be obtained. c. it will require large, unmanageable sample sizes. d. sample sizes are too small for most methods of statistical analysis. ANS: A A B C D Feedback Nonprobability sampling reduces or limits generalizability. Informed consent is used for all studies despite choice of sample used. Nonprobability sampling does not mandate large unmanageable sample sizes. Nonprobability sampling does not mandate a small sample size. 21. Which type of sampling is most at risk for sample bias? a. Quota b. Random c. Purposive d. Convenience ANS: D A B C D Feedback Quota sampling, if done appropriately, will not create sample bias. Random sampling should act to counteract sample bias. Purposive sampling would selectively include research participants and avoid sampling bias. Convenience sampling is most at risk for sample bias. 22. The primary characteristic of a probability sample is considered to be what? a. Self-selection of participants b. Random selection of the sample c. Participants handpicked by the researcher d. Representation of proportional segments of the population ANS: B A B C D Feedback Self-selection of participants would be antithetical to probability sampling. The primary characteristic of a probability sample is random selection. Participants being handpicked by a researcher represents a purposive sample, which is not a form of probability sampling. Proportional segment representation can be found in quota sampling, which is a form of nonprobability sampling. 23. A nurse researcher opts to use a stratified sample in a research study for which reason? a. The sample population is divided into subsets that are homogeneous for a particular trait or feature. The sample population is divided into subsets for random assignment into an intervention or a usual-care group. c. The sample population is randomly assigned to a specific subgroup that will vary from the others on the basis of type of treatment. d. The sample population is randomly assigned to a specific subgroup that will vary from the others on the basis of treatment duration. b. ANS: A A B C D Feedback The major characteristic of a stratified sample is that selected participants represent population subgroups that are homogeneous. A stratified sample would have representation in both the treatment and control groups. In a stratified sample, both treatment and control groups have representative groups. Stratified samples are not devised on the basis of length of treatment but on proportions of participants in the population. 24. What process is employed when a researcher accrues a sample of participants by using the word-of-mouth approach? a. Referral b. Networking c. Recruitment d. Quasiprobability ANS: B A B C D Feedback Referral is a generic term that is not associated specifically with the word-of-mouth approach to sampling. Networking is a sampling strategy that uses the word-of-mouth approach via social networks. The term recruitment describes the methods used to obtain research participants. The term quasiprobability is not associated with the word-of-mouth approach to sampling. 25. What does the technique of matching in a sampling strategy contribute to the research? a. Improvement of study design b. Elimination of the need for a control group c. Increased equivalency of the comparison group d. Reduction of sample size by making participants serve as their own controls ANS: C A B C D Feedback Matching, by itself, will not improve study design. Matching does not eliminate the control group; rather, it helps make the control group and the treatment group equivalent. Matching is a special strategy used to construct an equivalent comparison group. Matching does not reduce the sample size, and study participants cannot serve as their own controls. 26. A nurse researcher can most accurately use the technique of power analysis to do what? a. To estimate sample size b. To establish sampling strata c. To assign participants randomly d. To analyze sample representativeness ANS: A A B C D Feedback Power analysis is a statistical procedure used to determine needed sample size. Sampling strata would be determined by knowledge of population strata. A table of random numbers (or a computer program) can be used to assign participants randomly. Sample representativeness is not determined by power analysis. 27. A nurse researcher should be particularly concerned about having a small sample size for which reason? a. There is increased risk of obtaining a nonrepresentative sample. b. There is increased risk of a study participant interacting with another study c. d. participant. There is increased potential for loss of confidentiality. There is increased regression toward the mean. ANS: A A B C D Feedback Small sample sizes tend to increase the risk of a sample being nonrepresentative. Study participant interaction can occur, whatever the sample size. Confidentiality loss can affect large or small samples. Regression toward the mean is not generated by sample size. 28. What question should a research consumer ask when evaluating the sample size in a research report? a. How many assistants were involved in data collection? b. How representative is the sample relative to the target population? c. How great is the outcome difference between or among study groups? d. How many previous studies has the researcher conducted before on similar topics? ANS: B A B C D Feedback The number of data collectors is not of critical importance in research report evaluation. A major factor to be evaluated about sample size is how representative the sample is of the target population. Outcome differences would have greater impact on later research but not the current study. The expertise of the researcher does not affect sample size. 29. Which of the following terms describes the process of obtaining the required number of participants by asking those people who are easiest to access? a. Selection bias b. Convenience c. Randomization d. Internal reliability ANS: B A B C D Feedback Selection bias occurs when precautions are not exercised to ensure getting a representative sample. The term convenience refers to the type of sample where available participants, who do not necessarily represent the population, are included. Randomization is the process of obtaining participants that ensures that every participant in the population has an equal chance of being selected. The term internal reliability describes a characteristic of measurement where processes are standardized or where items in a survey relate to one construct. 30. Which statement regarding sample size is most accurate? a. Studies with smaller sample sizes have more accurate results. b. Studies with smaller sample sizes are more likely to be representative of the target c. d. population. Studies with larger sample sizes have more accurate results. Studies with larger sample sizes are more likely to be representative of the target population. ANS: D A B C D Feedback Smaller sample sizes may be associated with less accurate results. Smaller sample sizes are more likely to be unrepresentative of the population. Sample size does not guarantee accuracy of the results. In general, studies with larger sample sizes are more representative of the population. MULTIPLE RESPONSE 1. What are types of nonprobability of sampling? (Select all that apply.) a. Purposive b. Quota c. Stratified random d. Multistage (cluster) e. Convenience ANS: A, B, E Feedback Correct Incorrect Purposive sampling is a nonprobability type. Quota sampling is a nonprobability type. Convenience sampling is a nonprobability type. Stratified random is probability sampling. Cluster sampling is probability sampling. 2. The nurse researcher is deciding on a sampling strategy for the research study. Which type(s) of strategies would be considered to be probability sampling strategies? (Select all that apply.) a. Convenience b. Quota c. Simple random d. Cluster e. Purposive f. Stratified random ANS: C, D, F Feedback Correct Incorrect Simple random, cluster, and stratified random are probability sampling strategies. Convenience, quota, and purposive are nonprobability sampling strategies. 3. The nurse researcher is attempting to develop a purposive sampling strategy as part of a research study. What criteria will be used to determine whether the strategy is purposive? (Select all that apply.) a. Validation of scale with a known-group technique b. Focus of study population relates to specific diagnosis c. Focus of study population relates to broad, general topic d. Effective posttesting of instruments e. Collection of exploratory data f. Collection of descriptive data ANS: A, B, E, F Feedback Correct Incorrect Validation of scale with a known-group technique; Focus of study population relates to specific diagnosis; Collection of exploratory data; and Collection of descriptive data are criteria used to determine whether the sampling strategy is purposive. The study population should relate to a specific diagnosis, not to a broad topic. The instrument should be pretested, not posttested.