Nursing Research Finals Exam 2024-2025 PDF

BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 DESIGN AND PLANNING PHASE - Frequency of educational sessions or...

BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 DESIGN AND PLANNING PHASE - Frequency of educational sessions or activities (e.g., weekly, monthly, etc.). - Estimated duration of each session or SELECTING A RESEARCH DESIGN activity. o Overall plan for obtaining answers to the IV. TARGET AUDIENCE research questions. o Clearly define the target population or audience o Quantitative designs tend to be structured and for the intervention, such as age groups, controlled, with the goal of minimizing bias. demographics, and any specific characteristics. o Architectural backbone of the study. o Explain how participants will be recruited or selected. DEVELOPING PROTOCOLS FOR THE V. INTERVENTION COMPONENTS INTERVENTION o Detail the key components of the educational o In experimental research, researchers introduce campaign, including: an intervention. - Educational materials (e.g., brochures, o An intervention protocol for the study must be pamphlets, videos). developed, specifying exactly what the - Workshops or training sessions. intervention will entail (e.g. who will administer - Interactive sessions or community it, over how long a period will the treatment last, events. and so on) and what the comparative condition VI. DELIVERY METHODS will be. o In nonexperimental research, this step is not o Describe how the intervention will be delivered, necessary. including the use of in-person sessions, online platforms, or a combination of both. INTERVENTION PROTOCOLS VII. EVALUATION AND DATA COLLECTION I. INTRODUCTION VIII. ETHICAL CONSIDERATIONS o Briefly explain the purpose of the intervention. o Provide a rationale for the study, emphasizing IX. STATISTICAL ANALYSIS the importance of responsible antibiotic use. X. CONCLUSION II. INTERVENTION DESCRIPTION o Describe the content and of format. o Who will be responsible for administering the IDENTIFYING THE POPULATION project? o For those conducting quantitative research, it's III. DURATION AND FREQUENCY essential to define the specific traits that participants should have. o Outline the timeline of the intervention, o In other words, they need to pinpoint the group specifying: they'll be studying. This group, known as the - Start and end dates of the study. population, consists of all individuals or items JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 that share certain common features (this refers DESCRIPTIVE STUDIES to the "P" element in PICO questions). o Phenomena are described or the relationships PICO QUESTIONS between variables are examined. o When enough information exists → descriptive I. POPULATION (P) studies may be conducted in which hypotheses o What is the demographic and health status of are tested. the target population in Plaridel, Bulacan in - For example, researchers might be terms of antibiotic use? interested in describing the relationship between fatigue and psychological II. INTERVENTION (I) distress in HIV patients. o What is the impact of the antibiotic-stewardship - Kim (2020) conducted a descriptive program-based educational campaign in correlational study of nurses from 27 Plaridel, Bulacan? Korean hospitals to examine o How does participation in the educational relationships among stress, emotional campaign affect the knowledge, practices, and labor strategies (e.g., regulation and attitudes of the population regarding antibiotic suppression of felt emotions), and use? burnout. III. COMPARISON (C) EXPLORATORY STUDIES o What are the differences in knowledge, o Conducted when little is known about the practices, and attitudes towards antibiotic use phenomenon of interest. between individuals who participated in the o Findings would add to the body of knowledge. educational campaign and those who did not? o A flexible approach rather than a structured approach to data collection. IV. OUTCOME (O) o Hypotheses are generally not appropriate. o What are the changes in knowledge levels - For example, a descriptive quantitative regarding antibiotics before and after the study of patients’ preoperative stress educational campaign? might document the degree of stress o How does participation in the educational patients feel before surgery and the campaign influence antibiotic usage practices? percentage of patients who are stressed. - What factors diminish or increase a patient’s stress? Are nurses’ behaviors related to a patient’s stress level? EXPLORATORY, DESCRIPTIVE AND - Qualitative methods are especially EXPLANATORY STUDIES useful. o The amount of existing knowledge about the EXPLANATORY STUDIES variable(s) can be used as the criterion for classifying research as exploratory, descriptive, o Search for causal explanations and are much or explanatory. more rigorous. o Usually experimental. o Enough knowledge exists about the variables. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 - Quantitative Example: Golfenshtein and PART 1 Drach001-Zahavy (2015) tested a QUANTITAIVE RESEARCH DESIGNS theoretical model (attribution theory) to understand the role of patients’ attributions in nurses’ regulation of emotions in pediatric hospital wards. EXPERIMENTAL DESIGNS - Qualitative Example: Smith-Young and o Experimental Research → is concerned with collegues (2014) conducted an in-depth cause-and-effect relationships. study to develop a theoretical o Cause-and-Effect Relationship → occurs when understanding of the process of one object or event makes some other object or managing work-related musculoskeletal event happen. disorders while remaining at the o All experimental studies involve manipulation workplace. They called this process or control of the independent variable (cause) constant negotiation. and measurement of the dependent variable (effect). o In an experiment (or randomized controlled RESEARCH METHODOLOGY trial, RCT), researchers are active agents. o The controlled experiment is the gold standard o Considered to be the entire strategy for the for yielding reliable evidence about causes and study (from identification of the problem to final effects. plans for data collection). – Fain (2013). o It discusses the steps and procedures that you Validity of Experimental Design will use in conducting your research. o Extraneous Variables (Confounding/Intervening Variables) → variables the researcher is unable to control, or RESEARCH DESIGN does not choose to control, which can influence o Outline/ plan for a research study. the results of a study. o The blueprint for conducting a study (Fain, - The researcher acknowledges these 2013). study limitations in the discussion o This section describes the specific research section. approach you will use in the study (quantitative - In experimental studies, the extraneous or qualitative? experimental? phenomenology? variables, or competing explanations for etc.). the results, are labeled threats to internal o Concerned with the type of data that will be and external validity (Campbell & collected, and the means used to obtain the Stanley, 1963). needed data. o Internal Validity → the degree to which o Must be appropriate to test the study changes in the DV (effect) can be attributed to hypothesis(es) or answer the research the IV (cause). question(s). - Threats to internal validity are factors other than the IV that influence the DV. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 - These factors constitute rival o How to control? A comparison group of similar explanations or competing hypotheses children helps control for this threat. that might explain the study results. 4. Testing o External Validity → the degree to which study results can be generalized to other people and o Testing threat may occur in studies where a other settings. pretest is given or where subjects have - With what degree of confidence can the knowledge of baseline data. study findings be transferred from the o Testing refers to the influence of the pretest or sample to the entire population? knowledge of baseline data on the posttest - Will these study findings hold true with scores. Subjects may remember the answers they other groups in other times and places? put on the pretest and put the same answers on o As the researcher attempts to control for internal the posttest. validity, external validity is usually decreased. 5. Instrumentation Change o The researcher must decide how to balance internal and external validity. o Considered when mechanical instruments or judges are used in the pretest and posttest. Threats to Internal Validity o Instrumentation change involves the difference 1. Selection Bias between the pretest and posttest measurement caused by a change in the accuracy of the o Threat occurs when study results are attributed instrument or the judges’ ratings, rather than to the experimental treatment or the researcher’s because of the experimental treatment. manipulation of the IV when, in fact, the results o How to control? Training sessions for judges are related to subject differences before the IV and trial runs to check for fatigue factors may was manipulated. help control for instrumentation changes. If o How to control? Randomly assign subjects to mechanical instruments are used, these experimental and comparison groups. instruments should be checked for their 2. History accuracy throughout the study. o Occurs when some event besides the 6. Mortality experimental treatment occurs during a study, o Occurs when the subjects do not complete a and this event influences the DV. study; Attrition or dropout may occur in any o How to control? inclusion of at least one research study. simultaneous control or comparison group in a o The observed effects may occur because the study. Random assignment of subjects to subjects who dropped out of a particular group groups. are different from those who remained in the 3. Maturation study. o Note: For ethical reasons, participants can never o Becomes a threat when changes that occur be forced to remain in a study. within the subjects during an experimental o Hudson and Llosa (2015) suggested that subjects study influence the study results. drop out of studies due to lack of interest and o People may become older, taller, or sleepier motivation. from the time of the pretest to the posttest. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 o How to control? Establish a relationship with R = random assignment of subjects to groups the study participants and help them recognize X = experimental treatment or intervention the importance of their continued participation. O = observation or measurement of dependent variable Threats to External Validity o X’s and O’s on one line apply to a specific 1. Hawthorne Effect group. o The time sequence of events is read from left to o Effect occurs when study participants respond right. in a certain manner because they are aware that o If an X appears first and then an O, this means they are being observed. the intervention occurred first then an o This term came about as the result of the studies observation was made. on worker productivity at the Hawthorne plant o If a subscript appears after an X or O (X1; X2; O1; of the Western Electric Company. O2), the numbers indicate the first treatment, o How to control? By using a double-blind second treatment, first observation, second experiment neither the researcher nor the observation, and so forth. research participants are aware of which - Example: participants are in the experimental group and R O1 X O2 (Experimental Group) which participants are in the control group R O1 O2 (Comparison Group) 2. Experimenter Effect - This example has two groups, both of which were formed through random o Occurs when researcher characteristics or assignment (R) of subjects to groups. behaviors influence subject behaviors (facial - Both groups in the example were expression, age, gender, etc.). measured or given a pretest (O1) on the o Experimenter effect - experimental research phenomenon of interest (DV). The o Rosenthal effect - nonexperimental studies experimental group was exposed to an - Used to indicate the influence of an experimental treatment (IV); the interviewer on respondents’ answers. comparison group was not exposed to Researcher characteristics such as this treatment. Then, both groups were gender, dress, and type of jewelry may again measured or given the posttest influence study participants’ answers to (O2) on the phenomenon of interest questions in nonexperimental studies (DV). 3. Reactive Effects of the Pretest I. TRUE EXPERIMENTAL DESIGNS o occur when subjects have been sensitized to the o Those in which the researcher has a great deal of treatment because they took the pretest. This control over the research situation. sensitization may affect the posttest results. o Threats to the internal validity of the study are o People might not respond to the treatment in the minimized. same manner if they had not received a pretest. o Only with the use of true experimental designs Symbolic Presentation of Research Designs may causality be inferred with any degree of confidence. o Based on the notation scheme of Campbell and Stanley (1963). JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 Three Criteria of True Experimental Design o Although randomization is the preferred method for equalizing groups, there is no 1. Manipulation guarantee that the groups will be equal. o The researcher manipulates the experimental o Random assignment versus random sampling. variable(s). - Random Assignment → is a signature o First criterion for a true experimental design is of an experimental design (RCT). If manipulation of the IV. subjects are not randomly assigned to o It involves doing something to study intervention groups, then the design is participants (Polit and Beck, 2018). not a true experiment. o Experimenters manipulate the IV by - Random Sampling → refers to a administering a treatment (intervention) to some method of selecting people for a study. people and withholding it from others or Random sampling is not a signature of administering a different treatment. an experimental design. In fact, most RCTs do not involve random sampling. 2. Comparison/Control Group I.I. PRETEST-POSTTEST CONTROL GROUP o At least one experimental and one comparison DESIGN group are included in the study. o The second criterion for a true experimental o “Before-and-After Design”, the most frequently design is the use of a comparison or control used experimental design. group. o In this design: o A control group usually indicates a group in an - (a) the subjects are randomly assigned to experimental study that does not receive the groups, experimental treatment. - (b) a pretest is given to both groups, o In nursing research, the withholding of a - (c) the experimental group receives the treatment may be unethical. experimental treatment, and the comparison group receives the routine 3. Random Assignment treatment or no treatment, and o Subjects are randomly assigned to either the - (d) a posttest is given to both groups. experimental or the comparison group. R O1 X O2 (Experimental Group) o The third criterion for true experimental studies R O1 O2 (Comparison Group) is the random assignment of subjects to groups. o The researcher can determine if the groups were o Random assignment ensures that each subject equal before the treatment was administered. has an equal chance of being placed into any of o If the groups were not equivalent, the posttest the groups in an experimental study. scores may be adjusted statistically to control for o Researchers typically either use computers to the initial differences between the two groups perform the randomization. that were reflected in the pretest scores. o Random assignment concerns the equality of o This design controls for all threats to internal groups in experimental studies. The random validity. assignment of subjects to groups eliminates o Disadvantage: The external threat of the reactive selection bias as a threat to the internal validity effects of the pretest. of the study. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 I.II. POSTTEST-ONLY CONTROL GROUP DESIGN o Considered to be the most prestigious o Also known as after-only design: experimental design because it minimizes - (a) subjects are randomly assigned to threats to internal and external validity. groups, o This design not only controls for all the threats - (b) the experimental group receives the to internal validity, but also controls for the experimental treatment, and the reactive effects of the pretest comparison group receives the routine treatment or no treatment, and Advantages of Experiments - (c) a posttest is given to both groups. o RCTs are the “gold standard” for intervention R X O1 (Experimental Group) studies (therapy questions) because they yield R O1 (Comparison Group) the most persuasive evidence about the effects of o Some researchers argue that pretesting is not an intervention. always necessary, especially when o The great strength of experiments lies in the randomization is used. confidence with which causal relationships can o Easier to carry out and superior to the pretest- be inferred. Through the controls imposed by posttest design. intervening, comparing, and—especially— o The researcher does not have to be concerned randomizing, alternative explanations can often with the reactive effects of the pretest on the be ruled out posttest. o This design should be used when it is not Disadvantages of Experiments possible to administer a pretest or when it o Many interesting variables simply are not would not make sense to administer a pretest. amenable to intervention. I.III. SOLOMON FOUR GROUP DESIGN o Many variables could technically—but not ethically—be experimentally varied. - (a) subjects are randomly assigned to o Sometimes, RCTs are not feasible because of one of the four groups, practical issues. - (b) two of the groups, experimental group 1 and comparison group 1, are Note: How can you tell if a study is experimental? If the pretested, article says that the study purpose was to test the effects - (c) two of the groups, experimental of an intervention, and if participants were put into group 1 and experimental group 2, groups at random. receive the experimental treatment, II. QUASI-EXPERIMENTAL DESIGNS whereas two of the groups, comparison group 1 and comparison group 2, o Also called trials without randomization in the receive the routine treatment or no medical literature, are those in which there is treatment, and either no comparison group or subjects are not - (d) a posttest is given to all four groups. randomly assigned to groups. R O1 X O2 (Experimental Group 1) o The researcher uses existing, or intact, groups R O1 O2 (Comparison Group 1) for the experimental and comparison groups. R X O2 (Experimental Group 2) o It involves an intervention (just like true R O2 (Comparison Group 2) experiments); however, quasi-experimental JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 designs lack randomization, the signature of a considered very weak and in which the true experiment. researcher has little control over the research. o Advantage: By conducting experiments with III.I. ONE SHOT CASE STUDY naturally occurring groups, the real world is more closely approximated than when subjects o A single group is exposed to an experimental are randomly assigned to groups (practicality). treatment and observed after the treatment. o Disadvantage: It is weaker because, without XO randomization, it cannot be assumed that the o A group of patients with diabetes might attend a experimental and comparison groups are diabetic education class (X) and be tested on equivalent at the outset. Causal inferences their knowledge of diabetes (O) after the class is cannot be made as readily as with RCTs. completed. II.I. NONEQUIVALENT CONTROL GROUP DESIGN o This design does not call for any comparisons to be made. o Is like the pretest-posttest control group design o Threats to internal validity: history, maturation, except there is no random assignment of subjects and selection bias. to the experimental and comparison groups. o The threats of testing and instrumentation O1 X O2 (Experimental Group) change would not be applicable in this design. O1 O2 (Comparison Group) o Weakest of all the experimental designs because o Threats to internal validity controlled are it controls for no threats to internal validity. history, testing, maturation, and III.II. ONE GROUP PRETEST-POSTTEST DESIGN instrumentation change. o The biggest threat to internal validity: selection o Provides a comparison between a group of bias. subjects before and after the experimental o Note: A statistical test called analysis of treatment. covariance (ANCOVA) can be used to help O1 X O2 control for differences that might have existed, o A group of patients with diabetes could be given through chance, between the experimental and a pretest of their diabetes knowledge (O1). This control groups at the beginning of the study. group would then attend a diabetic education II.II. TIME SERIES DESIGN class (X) and be post tested (O2) at the end of the class. o The researcher periodically observes or o Threats to internal validity: history, maturation, measures the subjects. The experimental testing, and instrumentation change. treatment is administered between two of the o Because of the existence of a pretest and posttest, observations. testing and instrumentation change now become O1 O2 O3 X O4 O5 O6 threats to internal validity. o The greatest threats to validity: history and testing. NONEXPERIMENTAL DESIGNS III. PRE-EXPERIMENTAL DESIGNS o Is the name applied by Campbell and Stanley o Termed as “observational” in medical literature, (1963) to experimental designs that are many times, experimental research cannot be JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 conducted with human beings because of ethical o A correlation indicates the extent to which one reasons. variable (X) is related to another variable (Y). As o All nonexperimental research is descriptive X increases, does Y increase or decrease? because there is no manipulation or control of o In a simple correlational study, one group of variables, and the researcher can describe the subjects is measured on two variables (X and Y) phenomenon only as it exists. to determine if there is a relationship between o Although the researcher cannot talk about a these variables. cause-and-effect relationship in o A famous research dictum: correlation does not nonexperimental research, it is important to prove causation. obtain valid study results in this type of research o The magnitude and direction of the relationship between two variables is indicated by a I. SURVEY STUDIES correlation coefficient. o Are investigations in which self-report data are - Correlation Coefficients → may be collected from samples with the purpose of positive (+) or negative (-) and range describing populations on some variable(s) of from -1.00 (perfect negative correlation) interest. to 1.00 (perfect positive correlation). If o Probability sampling techniques and adequate the correlation coefficient has no sign in sample sizes are very important in survey front of it (e.g.,.80), a positive research. relationship exists. A negative o A nurse researcher might use a survey to gather correlation coefficient is preceded by a data on the health needs of clients, their sleep negative sign (e.g., -.80). A correlation patterns, or their perceptions of the nursing care coefficient of.00 indicates there is no they have received. relationship between variables. o The most common data-collection techniques - Correlation Coefficients →are reported used in survey research are questionnaires and through various statistics such as the interviews. Pearson’s product-moment correlation o Advantages: Ability to provide accurate (more commonly called the Pearson r) information on populations while using and the Spearman rho. A positive relatively small samples. The large amount of relationship (or direct relationship) data that can be obtained rather quickly and means that as the value of one variable with minimal cost. increases, the value of the other variable o Disadvantage: Self-report responses may be increases. A negative relationship unreliable because people may provide socially (inverse relationship) means that as the acceptable responses. value of one variable increases, the value of the other variable decreases. II. CORRELATION STUDIES Identification of an independent and a o The researcher examines the strength of dependent variable may not be relationships between variables by determining appropriate in some correlational how changes in one variable are associated with studies. Example: “A correlation changes in another variable. between age (IV) and assertiveness levels DV)”. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 o Advantage: Can offer an efficient way to collect o Experimental studies are inherently prospective large amounts of data about a problem. (by because the researcher institutes the intervention doing this, many relationships could be and subsequently examines its effect. discovered in a short time) o It starts with the determination of a cause and o Disadvantages: They do not yield persuasive then looks forward in time to determine the evidence for causal inferences. This is not a effect on subjects. problem when the aim is description, but o Prospective studies may use an experimental correlational studies are often undertaken to approach, whereas retrospective studies would discover causes. Yet correlational studies are never use this type of design. susceptible to faulty interpretation because o In prospective studies, the researcher might groups being compared have formed through manipulate the IV, or the cause, then observe self-selection. study participants in the future for the DV, or the effect. III. COMPARATIVE STUDIES o Prospective studies are costly, and subject o Examine the differences between intact groups dropout may occur. These types of studies are on some dependent variable of interest. less common than retrospective studies. o There is no manipulation of the IV (this design is used when manipulation of the independent variable may not be possible). TIME DIMENSION IN RESEARCH DESIGN o Classified as: Retrospective or Prospective. I. CROSS-SECTIONAL DESIGNS III.I. RETROSPECTIVE STUDIES o Data are collected at one point in time. o The DV (effect) is identified in the present (a o Retrospective studies are usually cross-sectional disease condition, for example), and an attempt o Can be used to study time-related phenomena, is made to determine the IV (cause of the but they are less persuasive than longitudinal disease) that occurred in the past. designs. o Are frequently called ex post facto → data are o Cross-sectional designs are economical, but they collected after the fact. Variations in the IV are pose problems for inferring changes over time. studied after the variations have occurred, rather than at the time of the occurrence. II. LONGITUDINAL DESIGNS o A retrospective study starts by examining an o Involve collecting data multiple times over an effect and then looks back in time to determine extended period. the cause. o Such designs are useful for studying changes III.II. PROSPECTIVE STUDIES over time and for establishing the sequencing of phenomena, which is a criterion for inferring o Cohort Design, the IV or presumed cause (high causality. cholesterol blood levels, for example) is o In nursing research, longitudinal studies are identified at the present time, then subjects are often follow-up studies of a clinical population, followed in the future to observe the DV undertaken to assess the subsequent status of (incidence of coronary artery disease, for people with a specified condition or who example). received an intervention. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 o Researchers must decide the number of data collection points and the time intervals between them. o A challenge in longitudinal studies is the loss of participants (attrition) over time. PART 2 QUALITATIVE RESEARCH DESIGN I. PHENOMENOLOGY II. ETHNOGRAPHY III. GROUNDED THEORY IV. CASE STUDY V. NARRATIVE INQUIRY VI. HISTORICAL RESEARCH JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 POPULATION AND SAMPLING o Exclusion Criteria → features of a person that, if met, immediately deems that person ineligible to participate in the study. Qualities about a OVERVIEW person or the external factors around them that would go against the goal of your study or o Quantitative Research: (Goal) would be to interfere with it. conduct a random sampling that ensures the sample group would be representative of the entire population (results could be generalized TYPES AND SUBTYPES OF SAMPLING to the entire population). o Qualitative Research: (Goal) is to provide in- I. PROBABILITY SAMPLING depth understanding and therefore, targets a o Scientific/Unbiased/Random Sampling specific group, type of individual, event or o Mostly used in quantitative research. process. o Goal: To obtain representative elements of populations. o All the members of an entire population have a TERMS TO REMEMBER chance of being selected. o Population → is a complete set (entire) of I.I. SIMPLE RANDOM SAMPLING persons/ objectives that possess some common characteristic of interest to the researcher. o All the members of the population are given an o Sample → subset of the population, selected to equal chance of being selected. represent the population; Method of selection o It is an unbiased way of selection, as samples are and Sample size→ will determine how drawn by chance, best type of probability representative a sample of the population. sampling. o Element → a single member of a population. o The only basis of including or excluding a o Sample Frame → elements/ members of member is by chance or opportunity. population are selected from here. It’s a listing o Advantages: Little knowledge of the population of all elements of a population (e.g. membership is needed, most unbiased of probability list). methods, and easy to analyze data and compute o Sampling → method or process of selecting errors. respondents or people to answer questions o Disadvantages: A complete listing of population, meant to yield data for a research study. time consuming, and expensive (esp. if with o Target Population → is composed of the entire large samples). group of people or objects to which the o Various ways of obtaining samples (Treece & researcher wishes to generalize the findings of Treece, 1986): the study. - Roulette wheel, o Accessible Population → a portion of the - Fishbowl method, and population to which the researcher has - Table of random number. reasonable access. Fishbowl Method o Inclusion Criteria → different requirements someone must meet to participate in your study. o Observes the following steps: JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 - The sampling frame, or a list of all the I.III. CLUSTER SAMPLING subjects or elements in the population in o Used in large-scale studies, where the question, should be prepared. population is geographically spread out. - All the names of the subjects or elements o The groups are defined to maintain the should be written down on strips of heterogeneity of the population. paper (one name per strip of paper). o Sampling technique in which the population is - The strips of paper with the listed divided into already existing groupings names are then placed in a bowl or (clusters), and then a sample of the cluster is container. selected randomly from the population. - Samples can be drawn as desired. I.IV. SYSTEMATIC SAMPLING Table of Random Numbers o The researcher selects every nth member after o Observe the following steps: randomly selecting the first, through nth - A number should be assigned to each element as starting point. element of the accessible population. o Procedure: - The table of random numbers should be - Obtain a list of the total population (N). entered at an arbitrary or random - Sample size is determined (n). starting point. This can be done by - Sampling interval width (K) is closing one’s eyes and using a pen to determined by N/n. For instance, if the point at a number. researcher were seeking a sample of 50 - A systematic movement should be from a population of 500, the sampling followed for going up or down, left or interval would be: K (width) = 500/ 50= right, or diagonally. The direction is not 10. important but should be decided before - Every 10th element of the population list starting the process. This direction shall would be selected for the sample. continue until the total sample is o Advantages: Easy to draw sample, economical, selected. and time-saving technique. - Continue to select numbers until the o Disadvantages: Samples may be biased if desired sample is reached. ordering of population is not random and after I.II. STRATIFIED RANDOM SAMPLING the first element is chosen, population members no longer have equal chance of being chosen. o The population is first divided into different strata, and then the sampling follows. II. NONPROBABILITY SAMPLING o Age, gender, and educational qualifications are o Process of selecting respondents in which not all some possible criteria used to divide a members of the entire population are given a population into strata. chance of being selected as samples. o Advantages: Increase probability of sample o Also called non-scientific sampling and is being representative and assures adequate commonly used in qualitative research. number of cases for subgroups. o The subjects are chosen based on their o Disadvantages: Requires accurate knowledge of availability or the purpose of the study, and in population, may be costly to prepare stratified some cases, on the sole discretion of the lists, and statistics are more complicated. researcher. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 II.I. CONVENIENCE SAMPLING o Involves the assistance of subjects to help obtain other potential subjects. o Accidental or incidental sampling. o This method often helps researchers find and o Involves choosing respondents at the recruit participants that may otherwise be hard convenience of the researcher. to reach. II.II. QUOTA SAMPLING o Involves dividing the population into SAMPLE SIZE homogenous strata and selecting sample elements from each of these strata. QUANTITATIVE o It obtains members through convenience sampling. Factors to Consider in Determining the Sample Size o Quota → arises from the establishment of a 1. Homogeneity of the population → the higher the desired quota/ proportion for some population degree of homogeneity of the population, the smaller the variable of interest. sample size that can be utilized. o This is commonly done to ensure the inclusion of a particular segment of the population. 2. Degree of precision desired by the researcher → the larger the sample size, the higher the precision or II.III. PURPOSEFUL/PURPOSIVE SAMPLING accuracy of the results will be. o Also called judgmental sampling. 3. Types of sampling procedure → probability sampling o The most common sampling strategy. uses smaller sample sizes than nonprobability sampling. o The researcher uses his/her own expert judgement. Various Approaches to Determining the Sample Size o Involves handpicking subjects, usually to suit 1. Sample sizes as small as 30 are generally adequate to very specific intentions. ensure that the sampling distribution of the mean will o Participants are selected or sought based on approximate the normal curve (Shott, 1990). preselected criteria based on the research question. 2. When the total population is equal to or less than 100, o Neuman (2007) says that purposive sampling is this same number may serve as the sample size. This is appropriate in three situations: called universal sampling. - a) A researcher uses it to select unique 3. According to Gay (1976), the following are the cases that are especially informative, acceptable sizes for different types of research: - b.) A researcher may use purposive sampling to select members of a - Descriptive research– 10% to 20% may be difficult-to-reach specialized population, required and - Comparative research– 15 subjects or groups - c) A researcher wants to identify types 4. Slovin’s formula is used to compute for sample size of cases for in-depth investigation. (Sevilla, 2003). II.IV. SNOWBALL SAMPLING o Also known as chain referral sampling/ friendship pyramiding. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 QUALITATIVE for each gender and 3 age groups for each gender. Three Considerations Ethnographic Surveys Select a large and 1. The saturation or redundancy of the sample size must representative sample be considered. (purposeful or random based on purpose) with 2. The size of a sample is important to determine the numbers similar to those difference within a target population. in a quantitative study. 3. Approximation of a sample size grounded on the method of the study may be used. o Note: Once a sampling method has been determined, the researcher must consider the Rule of Thumb Based Approach sample size. Research Approach Rule of Thumb o In qualitative studies, sampling typically Case Study Select one case or one continues until information redundancy or person. saturation occurs. This is the point at which no Phenomenology Assess 10 people. If new information is emerging in the data. you reach saturation Therefore, in qualitative studies it is critical that prior to assessing 10 data collection and analysis are occurring people, you may use simultaneously so that the researcher will know fewer. when the saturation point is reached. Grounded Assess 20-30 people, Theory/Ethnography/Action which typically is Research enough to reach saturation. Rule of Thumb Based on Data Collection Data Collection Method Rule of Thumb Interviewing Key Interview approximately Informants 5 people. In-depth Interviews Interview approximately 30 people. Focus Groups Create groups that averages 5-10 people each. In addition, consider the number of focus groups you need based on groupings represented in the research question. That is, when studying males and females of 3 different age groupings, plan 6 focus groups, giving you one JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 MEASUREMENT o Frequency distributions and percentages are used with this type of data as well as some o The process of assigning numbers to variables. statistical tests. o Used in research, implies the quantification of information (numbers are assigned to the data). III. INTERVAL MEASUREMENT Advantages of Measurement o Interval data consist of real numbers. o Interval level of measurement concerns data that o It removes guesswork and ambiguity in not only can be placed in categories and ranked, gathering and communicating information. but also the distance between the ranks can be o It helps obtain reasonably precise information. specified. o Measurement is a language of communication. o The categories in interval data are the actual numbers on the scale, such as on a thermometer. LEVEL OF MEASUREMENT IV. RATIO MEASUREMENT I. NOMINAL MEASUREMENT o Considered the highest or most precise level of data. o Objects or events are named or categorized o Includes data that can be categorized and (mutually exclusive and exhaustive categories). ranked; in addition, the distance between ranks o Numbers are obtained through counting the can be specified, and what is referred to as a true frequency or percentages of objects or events in or natural zero point can be identified. The zero each of the categories. point on the ratio scale means there is a total o E.g. gender, religious affiliation, marital status, absence of the quantity being measured. and political party membership. o The amount of money in your bank account o These types of variables are frequently assigned could be considered ratio data because it is numbers on questionnaires, such as 0 for males possible to be zero. and 1 for females. These numbers are only symbols used for data analysis purposes and have no quantitative meaning. o Considered the lowest level or least rigorous of the measurement levels. II. ORDINAL MEASUREMENT o Data that can be rank ordered as well as placed Converting Data into categories. o Data can always be converted from one level to a o The exact differences between the ranks cannot lower level of measurement, but not to a higher be specified. level. o The numbers obtained from this measurement o Example: The number of requests by patients for process indicate the order rather than the exact pain medication could be converted to ordinal quantity of the variables. data. Requests could be categorized as follows: o E.g. anxiety levels of people in a therapy group more than 10 requests per day, 5–10 requests per might be categorized as mild, moderate, and day, and 0–4 requests per day. It would also be severe. possible to change these data to nominal data by JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 classifying the subjects into two groups: those RELIABILITY OF THE INSTRUMENT with “no requests per day” and those with “1 or more requests per day.” o The reliability of an instrument concerns its consistency and stability. Appropriate Level of Measurement o It is the consistency with which an instrument measures the attribute (variable). o The main considerations in determining the o Reliability needs to be determined whether the level of measurement for data are: instrument is a mechanical device, a written - The level of measurement appropriate questionnaire, or a human observer. for the type of data that are being o A correlation coefficient is determined between sought. Some variables, by their very two sets of scores or between the ratings of two nature, can be measured at only one judges: The higher the correlation coefficient, the level. E.g. gender can be measured at more reliable is the instrument or the ratings only a nominal level. from the judges. - The degree of precision desired when it o Correlation coefficients can range between-1.00 is possible to consider the data at more and +1.00, correlation coefficients computed to than one level of measurement. If test the reliability of an instrument are expected concerned about the precision of the to be positive correlations. data, the interval or ratio level of o Reliability must continually be assessed as the measurement. If ranked or categorized instrument is used with different subjects and data will be sufficient= ordinal data may under different environmental conditions. be used. If categories of data are all that is called for, nominal data will be I. STABILITY appropriate. o Stability of an instrument refers to its consistency over time. o Assessments of stability: Made through test– DATA COLLECTION INSTRUMENT retest reliability procedures. o Research instruments (research tools), are the o Researchers administer the same measure to a devices used to collect data. sample twice and then compare the scores. o The instrument facilitates the observation and o A physiological instrument, such as a measurement of the variables of interest. thermometer, should be very stable and o The type of instrument used in a study is accurate. determined by the data-collection method(s) o Questionnaires can also be checked for their selected. stability. A questionnaire might be administered to a group of people, and, after a time, the Note: Great care should be taken to select the most instrument would again be administered to the appropriate instrument(s). same people. Criteria for selection of the Instrument o The interval between the two testing periods may vary from a few days to several months or o Practicality → concerns its cost and even longer. appropriateness for the study population. o Stability over time (test-retest reliability)- may o Reliability of the instrument not be the appropriate for a research instrument o Validity of the instrument. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 when you anticipate a change among the VALIDITY OF THE INSTRUMENT respondents over time. o Concerns its ability to gather the data that it is Note: Reliability higher than 0.70 are often considered intended to gather content of the instrument is satisfactory, but coefficients greater than 0.80 are far of prime importance. preferable. o The most important characteristic of an instrument. II. EQUIVALENCE o The greater the validity of an instrument, the o Degree to which two different forms of an more confidence you can have that the instrument obtain the same results or two or instrument will obtain data that will answer the more observers using a single instrument obtain research questions or test the research the same results. hypotheses. - Alternate or Parallel Forms Reliability o May be established using a panel of experts, or → when two forms of the same an examination of the existing literature on the instrument are compared. topic. Statistical procedures, therefore, may not - Interrater (or interobserver) Reliability always be used in trying to establish validity as → involve having two or more trained they are when trying to establish reliability (if observers or coders make simultaneous, ever, correlational). independent observations. I. FACE VALIDITY III. INTERNAL CONSISTENCY o When a preliminary examination shows that it is o Or Scale Homogeneity, addresses the extent to measuring what it is supposed to measure (an which all items on an instrument measure the adequate means of obtaining the data needed). same variable. o Can be examined using experts in the content o It is concerned with the sample of items used to area or using individuals who have measure the variable of interest. characteristics like those of the potential o Example: If an instrument is supposed to research participants. measure depression, all the items on the o Subjective in nature; this type of validity is instrument must consistently measure rarely used alone. depression. o Also known as logical validity. o Usually evaluated by calculating coefficient o The procedure calls only for intuitive judgment. alpha (or Cronbach’s alpha). o Just by looking at the instrument, the researcher decides if it has face validity. It includes the font size, spacing, the size of the paper used, and other necessary details that will not distract respondents from answering the questionnaire (Kubiszyn and Borich, 2007). II. CONTENT VALIDITY o Concerned with the scope or range of items used to measure the variable. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 o Methods of evaluating the content validity of an the present time (concurrent) or predict instrument. participants’ responses in the future (predictive). - By comparing the content of the Types of Criterion Validity instrument with material available in the literature on the topic. 1. Concurrent Validity - Use of a panel of experts, a group of o Compares an instrument’s ability to obtain a people who have expertise in each measurement of participants’ behavior that is subject area. These experts (3 to 5) are comparable to some other criterion of that given copies of the instrument and the behavior. purpose and objectives of the study. o The degree of validity would be determined They then evaluate the instrument, through correlation of the results of the two tests usually individually rather than in a group. Comparisons are made between administered to several people. The correlation (validity) coefficient must be at least.70 to these evaluations, and the researcher consider that the two instruments are obtaining then determines if additions, deletions, similar data. or other changes need to be made. 2. Predictive Validity Note: An instrument’s content validity is necessarily based on judgment. No totally objective methods exist o Concerned with the ability of an instrument to for ensuring the adequate content coverage of an predict behavior or responses of subjects in the instrument, but it is increasingly common to use a panel future. of substantive experts to evaluate the content validity of o If the predictive validity of an instrument is new instruments. Researchers typically calculate a established, it can be used with confidence to content validity index (CVI) that indicates the extent of discriminate between people, at the present expert agreement. We have suggested a CVI value of.90 time, in relation to their future behavior. For as the standard for establishing excellence in a scale’s example, a researcher might be interested in content validity (Polit & Beck, 2006). knowing if a suicidal potential assessment tool - A third method is used when would be useful in predicting actual suicidal behavior in the future. knowledge tests are being developed. The researcher develops a test blueprint IV. CONSTRUCT VALIDITY designed around the objectives for the content being taught and the level of o The most difficult to measure. knowledge that is expected (e.g., o Concerned with the degree to which an retention, recall, and synthesis). instrument measures the construct it is supposed to measure. II. CRITERION VALIDITY o Construct - is a concept or abstraction created or constructed by the researcher. o Concerned with the extent to which an o Involves the measurement of a variable that is instrument corresponds to, or is correlated with, not directly observable, but rather is an abstract some criterion measure. o It assesses the ability of an instrument to concept derived from observable behavior. determine the research participants’ responses at Methods of Construct Validity JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 1. Known-groups Procedure type of validity maybe considered. The process is not always the same. o Instrument under consideration is administered to two groups of people whose responses are Interpretation of Reliability Coefficients expected to differ on the variable of interest. o Reliability coefficients are important indicators o For example, if you were developing an of an instrument’s quality. instrument to measure depression, the theory o If data fail to support a hypothesis, one used to explain depression would indicate the possibility is that the instruments were types of behavior that would be expected in unreliable—not necessarily that the expected depressed people. relationships do not exist. Knowledge about an o If the tool was administered to a group of instrument’s reliability thus is critical in supposedly depressed subjects and to a group of interpreting research results, especially if supposedly happy subjects, you would expect research hypotheses are not supported. the two groups to score quite differently on the o Various things affect an instrument’s reliability. tool. If differences were not found, you might o E.g. reliability is related to sample heterogeneity. suspect that the instrument was not really The more homogeneous the sample (i.e., the measuring depression. more similar the scores), the lower the reliability 2. Factor Analysis coefficient will be. Instruments are designed to measure differences among those being o A method used to identify clusters of related measured, and if sample members are like one items on an instrument or scale. another, it is more difficult for the instrument to o This type of procedure helps the researcher discriminate reliably among those with varying determine whether the tool is measuring only degrees of the attribute. A depression scale will one construct or several constructs. be less reliable with a homeless group than with o Correlational procedures are used to determine a general sample. if items cluster together. o Longer scales (those with more items) tend to be RELATIONSHIP BETWENN RELIABILITY & more reliable than shorter ones. VALIDITY Note: Many psychosocial scales contain two or more o Both qualities are considered when selecting a subscales, each of which tap distinct, but related, research instrument. concepts (e.g., a measure of independent functioning o Reliability is usually considered first because it might include subscales for motor activities, is a necessary condition for validity. An communication, and socializing). The reliability of the instrument cannot be valid unless it is reliable. subscales is typically assessed and, if subscale scores are However, the reliability of an instrument tells summed for an overall score, the scale’s overall nothing about the degree of validity. In fact, an reliability would also be assessed. instrument can be very reliable and have low Interpretation of Validity validity. o Validity is often considered first in the o An instrument does not possess or lack validity; construction of an instrument. Face validity and it is a question of degree. content validity may be examined, then some o An instrument’s validity is not proved, type of reliability is considered. Next, another established, or verified but rather is supported to a greater or lesser extent by evidence. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 o Strictly speaking, researchers do not validate an instrument but rather an application of it. o A measure of anxiety may be valid for presurgical patients on the day of an operation but may not be valid for nursing students on the day of a test. Of course, some instruments may be valid for a wide range of uses with different types of samples, but each use requires new supporting evidence. The more evidence that can be gathered that an instrument is measuring what it is supposed to be measuring, the more confidence people will have in its validity. Note: In quantitative studies involving self-report or observational instruments, the research report usually provides validity and reliability information from an earlier study—often a study conducted by the person who developed the instrument. If the sample characteristics in the original study and the new study are similar, the citation provides valuable information about data quality in the new study. Ideally, researchers should also compute new reliability coefficients for the actual research sample. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 DATA GATHERING METHODS AND PROCEDURES o Quantitative Observation of Behavioral, Education Data → observation checklists and scales. DATA COLLECTION Records Review → Documents → photocopier, scanner, o time consuming, it is sometimes considered the image capture, materials used for organizing, digital most exciting part of research. image organizing software. o Data are the pieces of information or facts that III. Biophysiologic Measures are collected in research. o Data collection is a systematic process of o To assess important clinical variables. gathering observations or measurements; it o Biophysical Data → diagnostic tools and allows you to gain first-hand knowledge and machines, recording forms. original insights into your research problem. o Clinical nursing studies involve biophysiologic instruments both for creating independent variables (e.g., a biofeedback intervention) and for measuring dependent variables. THREE TYPES OF DATA I. SELF-REPORT DATA OVERVIEW o Patient-Reported Outcome (PRO) Data. o Participants’ responses to researchers’ questions, I. SELF-REPORTS/PATIENT-REPORTED such as in an interview; the most common data OUTCOMES (PRO) collection approach. o Survey → questionnaires, psychological tests, o Structured self-report methods are used when exams. researchers know in advance exactly what they o Interview → interview schedule/guide, audio- need to know and can frame appropriate recorder. questions to obtain the needed information. o Focus Group → focus group discussion o Structured self-report data are collected with a guidelines, interview schedule, audio recorder. formal, written document—an instrument: o Journaling → diary, journals, audio recorder, - Interview Schedule → when the questions video recorder, camera. are asked orally face-to-face or by telephone - Questionnaire → when respondents II. OBSERVATION complete the instrument themselves. o People’s behaviors and characteristics. o Advantages: Self-reports frequently yield o Laboratory Experiments on Biological and information that would be difficult or Material Samples → equipment/s used to impossible to gather by other means. generate data from the samples (microscope, o Disadvantages: The most serious issue concerns camera, thermometer, etc.) and the written form the validity and accuracy of self-reports: How to orderly store the collected data. can we be sure that respondents feel or act the o Observation for Qualitative Research → way they say they do? observation notes, reflective notes, Question Form methodological notes, reflexive notes, analytical notes. JUDITH HAZEL T. LISONDRA | 3A NURSING RESEARCH | FIRST SEMESTER  BULACAN STATE UNIVERISTY COLLEGE OF NURSING ACADEMIC YEAR 2024-2025 1. Closed-Ended (Fixed-Alternative) Questions → the o A scale is a device that assigns a numeric score response options are prespecified. Options may range to people along a continuum, like a scale for from a simple yes or no to complex expressions of measuring weight. opinion. Such questions ensure comparability of o Social psychological scales differentiate people responses and facilitate analysis. with different attitudes, perceptions, and psychological traits. - Advantages: Can be accomplished easily by the - One technique is the Likert scale (Rensis respondents; Data are easier to analyze. Likert), which consists of several - Disadvantages: Questions are more difficult to declarative statements (items) that construct; Researchers might omit potentially express a viewpoint on a topic. important responses; Some respondents object Respondents are asked to indicate how to choosing from alternatives that do not reflect much they agree or disagree with the their opinions precisely.