Cohort: A non-experimental design in which a defined group of people (a cohort) is followed overtime to study outcomes for subsets of the cohorts; also called a prospective design. P. 234 prospective (cohort) designs (studies that begin with a resumed cause and look forward in time for its effect. 2. Randomized controlled (trial): A full experimental test of an intervention, involving random assignment to treatment groups; sometimes, phase Ill of a full clinical trial.
Experiments (or randomized controlled trials [Rests]) Involve manipulation (the researcher manipulates the Independent variable by Introducing a treatment or Interventions control (Including use of a control group that Is not given the Intervention and represents the comparative contractually); and randomization or random assignment (with people allocated to experimental and control groups at random to arm groups that are comparable at the outset). P. 232 3. Factorial: (p . 14) experimental designs in which two or more independent variables are simultaneously manipulated, permitting a separate analysis of the main effects of the independent variables and their interaction. Terms (know definition and applicability) Contractually: chi 9 (p. 202) In a research context, a contractually Is what would have happened to the same people exposed to a causal factor If they simultaneously were not exposed to the causal factor. An effect represents the difference between hat actually did happen with the exposure and what would have happened without it.
This contractually model is an idealized conception that can never be realized, but it is a good model to keep in mind in designing a study to provide cause-and- effect evidence. Confounding: p. 177 The issue of contaminating factors?called confounding (or extraneous) variables. A variable that is extraneous to the research question and that confounds the relationship between the independent and dependent variables; confounding variables need to be controlled either in the research design or through statistical procedures. Causality: chi 9 (p. 01) cause & effect Placebo: chi 9 A placebo or extemporaneously presumed to have no therapeutic value; for example, In studies of the effectiveness of drugs, some patients get the experimental drug and others get an innocuous substance. Placebos are used to to participants. (There can, however, be placebo effects?changes in the dependent variable attributable to the placebo condition?because of participants’ expectations of benefits or harms). Factorial design: chi 9 (p. 214) When two or more independent variables are manipulated simultaneously and allow researchers to test both main effects and interaction effects.
Randomized groups: Hawthorne Effect: p. 216 “a placebo-type effect caused by people’s expectations. The term is derived from a set of experiments conducted at the Hawthorne plant of the Western Electric Corporation in which various environmental conditions, such as light and working hours, were varied to test their effects on worker productivity. Regardless of what change was introduced, that is, whether the light was made better or worse, productivity increased. Knowledge of being included in the study (not Just knowledge of being in a particular group) appears to have affected people’s behavior, thus obscuring the effect of the treatment. . Masking: (chi 9 p. 233 ) Blinding (or masking) is sometimes used devoid biases stemming from participants’ or research agents’ awareness of group status restudy hypotheses. Single-blind studies involvement of one group (e. G. , participants) amendable-blind studies involve masking of two groups (e. G. ,participants, investigators). 2. Blinding: (same as masking) Also called Masking; Sometimes used to avoid biases stemming from participants’ or research agents’ awareness of group status or study hypotheses. . Prospective study: prospective (cohort) signs(studies that begin with a presumed cause and look forward in time for its effect) a. Cause & Effect: Chi 9 p. 233 1. Switching replication design: p. 268 Replication studies are direct attempts to see if findings obtained in a study can be duplicated in another study. 2. Time series designs: In a time series design, there is no comparison group; information on the dependent variable is collected over a period of time before and after the intervention.
Time series designs are often used in single-subject (N-of-l) experiments. 3. Nonequivalent control group pretest-posters: The nonequivalent intro group pretest-posters design involves using a normalized comparison group and the collection of pre-treatment data so that initial group equivalence can be assessed. 4. Quasi-experimental: Quasi-experimental designs (controlled trials without randomization) involve an intervention but lack randomization. Strong quasi- experimental designs include features in support of causal inferences. 5. ) After-only (posters-only) design: An experimental design in which data are collected from subjects only after the intervention has been introduced; also called an after-only design. ) Before-after (pretest-posters) design: An experimental design in which data are collected from subjects both before and after introducing an intervention; also called a before-after design. O Pretest-posters Design: chi 9 p. 233 A posters-only (or after-only)design involves collecting data only after an intervention. In a pretest-posters(or before-after) design, data are collected both before and after the intervention, permitting an analysis of change. ) Factorial design: An experimental design in which two or more independent variables are independent variables and their interaction. D) Crossover (repeated measures) sign : An experimental design in which one group of subjects is exposed to more than one condition or treatment, preferably in random order. 6. Manipulation (of Variable) : p. 203 – factorial design p. 234 chi 1. Correlations Study: describe how phenomena are interrelated without invoking causal explanations. 2. Inebriate descriptive study: examine the frequency or average value of variables. 3.
Cohort Study: prospective (cohort) designs studies that begin with a presumed cause and look forward in time for its effect. 4. Path Analysis: which test causal models developed on the basis of theory 5. Triangulation (in qualitative studies) : The use of multiple methods to collect and interpret data about a phenomenon, so as to converge on an accurate representation of reality. Types of correlation studies: p. 224 Although correlation studies are inherently weaker than experimental studies in elucidating cause-and-effect relationships, different designs offer different degrees of supportive evidence. Designs for correlation studies include: (1) retrospective (case-control) designs (which begin with the outcome and look back in time for antecedent causes of “saneness” by comparing cases that have a disease r condition with controls who do not) (2) prospective (cohort)designs (studies that begin with a presumed cause and look forward in time for its effect); (3) natural experiments (in which a group is affected by a seemingly random event, such as a disaster); (4) path analytic studies (which test causal models developed on the basis of theory). The aim of descriptive correlation research is to describe relationships among variables rather than to support inferences of causality. Example on page 232 Case control: (see below) Retrospective: p. 224 Also called Case-Control designs; Begin with the outcome ND look back in time for antecedent causes of “saneness” by comparing cases that Prospective: Also called Cohort designs; Studies that begin with a presumed cause and look forward in time for its effect. Crossover: When people are exposed to more than one experimental condition, administered in a randomized order, and thus serve as their own controls.
CHAPTER – p. 236-256 Rigor and Validity in Quantitative Research Limitations of research designs (quantitative): p. 216 & see CHI 10 PPTP Controlling Intrinsic Source of Confounding Variability – p. 237 CHI 10 1. Stratification: chi 9 & 10th division of a sample off population into smaller units e. G. , males and females), typically to enhance representatives or to explore results for subgroups of people; used in both sampling and in allocation to treatment groups. (238) 2. Randomization: chi 9 & 10 p. 237 The researcher assigns participants too control or experimental condition on a random basis. E most effective method of controlling individual characteristics. The primary function of randomization is to secure comparable groups?that is, to equalize groups with respect to confounding variables. 3. Crossover design: (p. 215) A crossover design involves exposing the same people to more than one condition. This type of within-subjects design has the advantage of ensuring the highest possible equivalence among participants exposed to different conditions?the groups being compared are equal with respect to age, weight, health, and so on because they are composed of the same people. . Matching: (p. 238) Matching (also called pair matching) involves using information What is the best approach to help control extraneous variables? P. 2550 The issue of contaminating factors?called confounding (or extraneous) variables o The best control method is randomization to treatment conditions, which effectively intros all confounding variables?especially within the context of a crossover design. Four types of validity that affect the rigor of a quantitative study. Know types. – CHI 10 1.
Statistical conclusion validity: Statistical conclusion validity concerns the validity of inferences that there is an empirical relationship between variables (most often ,the presumed cause and the effect). 2. Internal validity: p. 255 Internal validity concerns inferences that out-comes were caused by the independent variable, rather than by factors extraneous to the research. Threats to internal validity include: p. 236 chi 10 1. Emperor ambiguity (lack of clarity about whether the pre-summed cause preceded the outcome), 2. Election (preexisting group differences), 3. History (the occurrence of events external to an independent variable that could affect outcomes), 4. Maturation (changes resulting from the passage of time), 5. Mortality (effects attributable to attrition), 6. Testing (effects of a pretest), 7. Instrumentation (changes in the way data are gathered). Internal validity can be enhanced through Judicious design decisions, but can also be addressed analytically (e. G. , through an analysis of selection or attrition biases).
When people withdraw from study, an intention-to-treat analysis (analyzing outcomes for all people in their original treatment conditions) is preferred to a per- protocol analysis (analyzing outcomes only for those who received the full treatment as assigned) for maintaining the integrity of randomization. 3. Construct validity: p. 255 concerns inferences from the particular exemplars of a study (e. G. , the specific treatments, outcomes, people, and settings) to the higher-order constructs that they are intended to represent.
The first step in fostering construct validity is a careful explication of those constructs. Threats to construct validity can occur if the personalization of a construct fails to incorporate all of the relevant characteristics of the construct or if it includes extraneous content. Examples of such threats include subject reactivity, researcher expectancies, novelty effects, compensatory effects, and treatment diffusion. 4. External validity- concerns whether inferences about observed relationships will hold over variations in persons, setting, time, or measures of the outcomes.
External validity, then, is about the generalization of causal inferences, and this is a critical once for research that aims to yield evidence for evidence-based nursing practice. External validity can be enhanced by selecting representative people, settings, and When is a study internally valid? Study validity concerns the extent to which appropriate inferences can be made. Threats to validity are reasons that an inference could be wrong. A key function of quantitative research design is to rule out validity threats by exercising various types of control.
Control over confounding participant characteristics is key to managing many validity threats. The best control method is randomization to retirement conditions, which effectively controls all confounding variables?especially within the context of a crossover design. CHAPTER 20: p. 487-514 Qualitative Research Design and Approaches CHI 20: Qualitative research traditions have their roots in: (p. 489) 1. Ethnocentric (roots Anthropology)- focuses on the culture of a group of people and relies on extensive fieldwork that usually includes participant observation and in-depth interviews with key informants.
Ethan-graphs strive to acquire an mimic (insider’s) perspective of a culture rather than antic (outsider’s) perspective. 2. Ethnomusicology (roots Sociology): seeks to discover how people make sense of their everyday activities and interpret their social worlds, so as to behave in socially acceptable ways. Within this tradition, researchers attempt to understand a social group’s norms and assumptions that are so deeply ingrained that immerse no longer think about the underlying reasons for their behaviors. 3. Hermeneutics (allied with Phenomenology): focuses on interpreting the meaning of experiences, rather than just describing them.
Types of grounded theory studies: (p. 498) Grounded theory aims to discover theoretical precepts grounded in the data. Grounded theory researchers try to account for people’s actions by focusing on the main concern that the behavior is designed to resolve. 1. Substantive theory is grounded in data on a specific substantive area, such as postpartum depression. It can serve as a springboard for- 2. Formal grounded theory, which is at a higher level of conceptualization and is abstract of time, place, and persons.
The goal of formal grounded theory is not to discover a new core variable but to develop a theory that goes beyond the substantive grounded theory and extends the general implications of the core variable. . Charisma’s constructivist grounded theory has emerged as a method to emphasize interpretive aspects in which the grounded theory is constructed from shared experiences and relationships between the researcher and study participants. O Qualitative description: p. 505 qualitative description is perhaps viewed as a “distributed residual category'(p. 82) that signals a “confederacy’ of diverse qualitative inquirers.
CHI 21 sampling in Qualitative Research: 1. Sampling Plan: The formal plan specifying a sampling method, a sample size, and procedures for recruiting subjects. 2. Data Saturation: The collection of qualitative data to the point where a sense of closure is attained because new data yield redundant information. 3. Transferability: (p. 530) The extent to which qualitative findings can be transferred to other settings or groups; one of several models of generalization. 4. Reflexivity: In qualitative studies, critical self-reflection about one’s own biases, preferences, and preconceptions. 5.
Descriptive correlation: 6. Triangulation: The use of multiple methods to collect and interpret data about a phenomenon, so as to con-verge on an accurate representation of reality. . Patient- centered intervention: An intervention tailored to meet individual needs or characteristics. Use of Analysis of covariance for statistical control p. 443 Controlling confounding variables. Various approaches can be used to control confounding variables, many of which require measuring those variables. For example, for analysis of covariance, variables that are statistically controlled must be measured. P. 255 When randomization is not possible, other control methods include statistical control to remove the effect of a confounding variable statistically (e. G. , through analysis of covariance). Statistical power refers to the ability to detect true relationships among variables. Adequate statistical power can be achieved in various ways, the most straightforward of which is to use a sufficiently large sample. When small samples are used, statistical power tends to be low, and the analyses may fail to show that the independent and dependent variables are related?even when they are.
Another aspect of a powerful design concerns how the independent variable is defined. Typical sample size in qualitative study: (for various types of studies I. E. Phenomenology, grounded theory) p. 529 1. Ethnographers make numerous impaling decisions, including not only whom to sample, but also what to sample (e. G. , activities, events, documents, artifacts); decision making is often aided by their key informants who serve as guides and interpreters of the culture. 2. Phenomenological typically work with a small sample of people (10 or fewer) who meet the criterion of having lived the experience under study. . Grounded theory researchers typically use theoretical sampling in which sampling decisions are guided in an ongoing fashion by the emerging theory. Samples of about 20 to 30 people are typical in grounded theory studies. Typical sample size in grounded theory (see above) Types of Samples: 1. Convenience; p. 2761529 qualitative Convenience sampling entails using the most conveniently available people as participants. A faculty member who distributes questionnaires to nursing students in a class is using a convenience sample.
The nurse who conducts a study of teenage risk taking at a local high school is also relying on a convenience sample. The problem with convenience sampling is that those who are available might be a typical of the population with regard to critical variables. Weakest form of sampling 1. . Snowball: (p. 276/ 516) also called network sampling or chain sampling- is a variant of convenience sampling. With this approach, early sample members (called seeds) are asked to refer other people whom et the eligibility criteria.
This sampling method is often used when the population is people with characteristics who might otherwise be difficult to identify (e. G. , people who are afraid of hospitals). Snowballing begins with a few eligible participants and then continues on the basis of participant referrals. 3. Purposive / Purposeful : (279) or Judgmental sampling uses researchers’ knowledge about the population to select sample members. Researchers might decide purposely to select people who are Judged to be typical of the population or particularly knowledgeable about the issues under study. 4.
Responding to numerous criticisms and to their own evolving conceptualizations, a fifth criterion that is more distinctively within the constructivist paradigm was added: authenticity (Cuba & Lincoln,1994). What is credibility in the- framework of quality criteria? P. 599 which refers to confidence in the truth value of the findings, is sometimes said to be the qualitative equivalent of internal validity. to the extent to which researchers fairly and faithfully show a range of different realities and convey the feeling tone of lives as they are lived.