Research Process 3 Research Design without Ref..docx
Transcript
Research Process 3, Research Design Objectives Define the term Research Design Identify three main types of Research Designs Discuss the two major Research Methodologies used in Research Describe the Population Selection and Sampling Methods in Nursing Research Introduction In the next major phase o...
Research Process 3, Research Design Objectives Define the term Research Design Identify three main types of Research Designs Discuss the two major Research Methodologies used in Research Describe the Population Selection and Sampling Methods in Nursing Research Introduction In the next major phase of a research project, the investigator must make a number of decisions about the methods to be used to address the research question and must carefully plan for the actual collection of data. Each methodological decision that the researcher makes during this phase has implications for the quality, integrity and interpretability of the results. Methodology refers to the nuts and bolts of how the research study is undertaken. The first of the elements to be studied is the research design. Design A set of instructions for the researcher; to gather and analyze data in certain ways that will control who and what are to be studied. Thus, the choice of design is made when the question is finalized. Research Design The research design is the overall plan for obtaining answers to the research questions or testing the research hypothesis, and for handling some of the difficulties encountered during the research process. The design normally specifies which of the various types of research approaches will be adopted. The design must be appropriate to test the study hypothesis or hypotheses or answer the research question or questions. Types of Research Design Because of the importance of the time dimension in designing research, studies are often categorized in terms of how they deal with time. Cross- Sectional Longitudinal Experimental Cross-sectional Design Used to examine groups of subjects in various stages of development simultaneously with the intent of inferring trends overtime. Those in which data collection is strictly in the present time. Cross-sectional research design is one in which researcher collects data at a particular point of time (one period of data collection). These studies are easier and more convenient to carry out For example, a researcher is interested in assessing the awareness of COVID among people of an area Here the researcher interacts only once to collect awareness related data from respondents Longitudinal Design Involves data collection at two or more points over an extended period. Longitudinal design examines changes in same subjects over an extended period. Experimental Design Involves manipulations, the researcher manipulates the independent variable (cause) by introducing treatment or intervention. Data gathering is periodic and over long periods. Styles of Research One, Descriptive used when the purpose of the study is to inquire about the prevailing conditions of events, objects or people. The method describes “what is” in relation to the variables under consideration. Descriptive Types: Exploratory Descriptive Design Descriptive Survey Design Correlational Design Comparative Design Case study Feasibility Study Two, Associational correlational causal comparative three, Intervention experimental quasi-experimental An intelligent reader, of research findings, should be able to recognize the design used in a study report and evaluate its suitability, for addressing the study’s purpose. Research Methodology Research Methodology is a way to systematically solve the research problem. It may be understood as a science of studying how research is done scientifically Types Of Research Method The goal of nursing research is the creation or discovery of knowledge that will enhance, nursing practice, education and administration. There are two major methodologies used in research, which originate from different philosophical perspectives and use different methods for collection and analysis of data. Quantitative and Qualitative Research Methodologies. Quantitative Method Emphasis is placed on the collection of numerical data, the summary of those data and the drawing of inferences from the data. In quantitative research the investigator seeks to explain what causes something to happen and makes predictions based on theories. The investigator researches a problem by a process of deduction, which in part is based on his or her perspective of the phenomenon being studied. Quantitative Research Variables are controlled and pieces of the phenomenon are explored. The data in this type of research are usually numbers that are collected and then analyzed. Qualitative Research Qualitative research is defined as a systematic collection and analysis of more subjective narrative material, using minimum imposed controls. Fewer subjects are needed in qualitative studies. They are chosen based on them having some knowledge of the phenomenon being studied. Qualitative Method Is based on words, feelings, emotions, sounds and other non-numerical and unquantifiable elements. It has been noted that “information is considered qualitative in nature if it cannot be analysed by means of mathematical techniques”. For example, if a researcher was to study the feelings of nurses regarding their supervisor's leadership styles, the researcher would interview nurses who are being supervised. This would assure the researcher that the subjects of the study experienced the phenomenon being studied. The success of a qualitative study relies heavily on the researcher’s ability to get trustworthy data. The researcher’s primary responsibility is looking and listening; two behaviours that are also important to nursing. It is here the researcher becomes the student and the subjects are the teachers. In the past it was called the “soft science” and not given the same significance as quantitative methods. However, over the past 10 – 20 years, qualitative studies have proven invaluable to health care delivery, patient care and nursing. The data, which are generally taped interviews or observations of behaviours, are gathered and transcribed verbatim. Trustworthy findings are presented after texts are analysed with skill, flexibility and creativity. Quantitative Research The emphasis is on collecting measurable information. Qualitative Research Information can only be loosely measured: the main issues can be identified but not specifically measured. Quantitative Research Data can be analyzed statistically. Qualitative Research Data cannot be statistically analyzed. Quantitative Research Data can be quickly collected so large samples can be used. Qualitative Research Data collection is more time consuming so uses smaller samples. Quantitative Research Data can be collected from a distance so can be collected from widely dispersed members of the population. Qualitative Research Data is usually collected face to face so collecting data from a widely dispersed sample is time consuming and expensive. Local samples tend to used. Quantitative Research Data collection tools are highly structured and are time consuming to develop. Qualitative Research Data collection tools are more loosely structured. Quantitative Research Once the tool is developed, data collection is relatively quick and cheap to collect and analyse. Qualitative Research Data collection and data analysis are time consuming and comparatively expensive. Quantitative Research The main forms of data collection are questionnaire surveys, highly structured observation schedules and analysis of records. Qualitative Research The main forms of data collection are individual interviews, focus groups and less structured observation. Quantitative Research Researcher uses tools, such as questionnaires or equipment to collect numerical data. Qualitative Research The Researcher is the data gathering instrument. Quantitative Research Objective – seeks precise measurement and analysis of target concepts, example uses survey, questionnaires Qualitative Research Subjective – individuals’ interpretation of event is important example uses participant observation, in-depth interviews, focus groups. Quantitative Research Quantitative data is more efficient, able to test hypotheses, but may miss contextual detail. Qualitative Research Qualitative data is more rich, time consuming and less able to be generalized. Quantitative Research All aspects of the study are carefully designed before data is collected. Qualitative Research The design emerges as the study unfolds. Quantitative research involves analysis of numerical data. Qualitative Research research involves analysis of data such as words (interviews), pictures (video) or objects (an artifact) Quantitative and Qualitative Research It is important to understand that quantitative and qualitative approaches are both valid research methodologies in nursing research. Although they proceed from different philosophical perspectives and use different methods of data collection and analysis, they both have their place in research. MIXED METHOD Data collection involves gathering both numeric (questionnaires) as well text information (interviews or focus groups). Represents both quantitative and qualitative information. Employs strategies to collect data either sequentially or simultaneously Research Settings The research design must specify the site and setting for the research study. Natural or field setting: uncontrolled in real life – Seen in descriptive or correlational studies Partially controlled setting: manipulated or modified by the researcher – Seen in correlational, quasi, or experimental studies Highly controlled setting: artificially constructed by researcher (that is, lab setting) – Seen in experimental studies SAMPLING Population Terms Population: entire set of cases that meet the specified criteria. Refers to the aggregate or totality of all the objects, subjects, or members that conform to a specific set of specifications. example, All the mentally ill clients with schizophrenia at the Bellevue Hospital Target population: the entire set of individuals or (elements) who (that) meet the sampling criteria. The accessible population is the population in research to which the researchers can apply their conclusions. This population is a subset of the target population and is also known as the study population. It is from the accessible population that researchers select their samples. General population encircles Target and Accessible population. And target population encircles Accessible population Population to be Studied and Data Collection The requirement of defining a population for a research project arises from the need to specify the group to which the results of a study can be applied. It is seldom possible to study an entire population, unless it is particularly small. Research studies, as a rule, use as subjects, only a small fraction of the population, referred to as a sample. Sampling Terms Sampling - a process of selecting subjects that are representatives of the population being studied. A process of selecting a portion of the population to represent the entire population. Strata: mutually exclusive segments of the population based on one or more characteristics. Sampling bias: refers to the systematic over-representation or underrepresentation of some segment of the population Sample frame: the technical name for the actual list of the elements from which the sample will be chosen. The sampling frame is the list from which the potential respondents are drawn Nursing Office Admission Appointment Book Must assess sampling frame errors Sample intervals: the standard distance between the elements chosen for the sample, example every nth client with Bipolar. Sampling error is probability of getting an un-representative sample by chance. Sample may be biased if not drawn accurately. Steps in Sampling Define study population Specify sampling frame and unit Specify sampling method Determine sample size Specify sampling plan Choose sample Before one selects actual subjects, it is essential to know what characteristics the sample should possess and the method of sampling. SAMPLING BREAKDOWN Who do you want to generalize? The theoretical population. What population can you get access to? The study population. How can you get access to them? The sample frame. Who is it you're studying? The sample. Reasons for Sampling Means by which predictions about behavioral outcomes can be made Resources (time, money) and workload / More economical, efficient and practical Efficient means of collecting data Gives results with known accuracy that can be calculated mathematically Types of Sampling Methods Probability Simple Random Stratified Random Systematic Random Cluster Random Random Assignment Types of Sampling Methods Non- Probability Sampling Convenience Snowball Quota Purposive Probability Sampling Probability sampling methods have been developed to increase the representativeness of the sample. In probability sampling every member (element) of the population has a probability higher than zero of being selected for the sample. The hallmark of probability sampling is the random selection of elements from the population. Random selection ensures that each and every element in the population has an equal and independent chance of being selected. Simple Random Sampling This is the basic component of probability sampling. The procedure for conducting a simple random sample exercise involves the use of a sample frame. If the sample frame is small, names can be written on slips of paper, placed in a container, mix well, numbers are drawn until the desired sample size is reached. However, for larger sample frames, tables and computer programs are available for selecting larger samples. Stratified Random Sampling Random sampling in which the population is first divided into two or more strata. The most common procedure for drawing a stratified random sample is to group those elements that belong to a stratum and select randomly the desired number of elements. Stratified random sampling is used in situations in which the researcher knows some of the variables in the population that are critical to achieving representativeness. Variables commonly used for stratification include: age gender ethnicity socioeconomic status diagnosis geographic region type of care Stratified Sampling……. Draw a sample from each stratum Cluster Sampling Requires successive random sampling units. The first unit to be sampled is large groupings, or cluster. For example, in selecting a sample of Nurses, the researcher must first draw a random sample of Hospitals A sample frame is developed that includes a list of all the parishes, capitals, institutions, or organization with which elements of the identified population would be linked. A random sample of these elements would then be used in the study. Two types of cluster sampling methods. One-stage sampling. All of the elements within selected clusters are included in the sample. Two-stage sampling. A subset of elements within selected clusters are randomly selected for inclusion in the sample. Stratified versus Cluster Stratified to ensure enough samples from subgroups and to lower sampling error Cluster primarily to reduce costs of gathering the data Form homogeneous groups when (stratif-fying), heterogeneous when clustering Systematic Sampling This involves the selection of the “Nth” case from some list or group, such as every 10th patient in hospital As described above, systematic sampling is an 'equal probability of selection' (EPS) design, because all elements have the same probability of selection (in the example given, one in ten). It is not 'simple random sampling' because different subsets of the same size have different selection probabilities - example the set {4,14,24,...,994} has a one-in-ten probability of selection, but the set {4,13,24,33,...} has zero probability of selection. Random Assignment A procedure used to assign subjects to treatment or control group randomly Random assignment used without random sampling decreases the risk of bias in the selection of groups. Advantages of Probability Sampling Reliable Eliminates biases Allows for estimation of sampling errors Disadvantage of Probability Sampling Expensive Inconvenient Sometimes unnecessary Non- Probability Sampling Non - probability sampling: selection of the sample units using nonrandom methods. Reduce representativeness. Convenience Snowball Quota Purposive Convenience Sampling Convenience sampling- also referred to as accidental or incidental sampling involves choosing readily available people or objects for a study. Is considered a poor approach because it provides little opportunity to control for biases Subjects are included in the study merely because they happen to at the right time – Use results that are easy to get Quota Sampling Uses convenience sampling, but with a strategy to ensure inclusion of subject types who are likely to be underrepresented in the convenience sample. Goal is to replicate the proportions of subgroups present in the population. Works better than convenience sampling to reduce bias. Purposive Sampling Sometimes referred to as judgmental sampling, involves the conscious selection by the researcher of certain subjects or elements to include in the study (example, good, bad, effective or ineffective). Snowball Sampling Also called network sampling: this term refers to a method that involves the assistance of study subjects to help obtain another potential subject. It is also another method of convenience sampling Snowball sampling is a recruitment technique in which research participants are asked to assist researchers in identifying other potential subjects. How It Works: Start Small: Begin with a single participant, often someone with unique access or knowledge about the group being studied. Expand Through Referrals: After interviewing the initial participant, ask them to refer others who meet the study criteria. Chain Effect: Each new participant refers to more individuals. This creates a ‘snowball effect’ that enlarges the sample size over time. Advantages of Non- probability Sampling Convenient Economical Often used when a sample or experts or a selection subjects based on known characteristics is needed, example purposive sampling. Disadvantage of Non-Probability Sampling Rarely represents target population Increases probability sampling errors Sample size Based on four factors Cost or budget Accuracy desired variance in population on variable of interest subgroup analysis planned Sampling Want to estimate a characteristic of population (population parameter). Estimate a corresponding sample statistic Sample must be representative of population on variable or variables of interest
