NURS 205: Unit 3 - Introduction to Quantitative Research

Variables in a Quantitative Design

- Dependent

- Independent

- Antecedent

- Extraneous

Antecedent Variable

Something that happens before the research study that impacts the outcome of the research study (Ex Age,

gender, socioeconomic status, and health status)

Extraneous Variable (Mediating Variable)

Things that happen within the study that may affect the study (Ex. Dog in the Hugh Jackman video)

Considerations to Increase Control in a Study

- Objectivity

- Accuracy

- Feasibility

- Homogenous sample

- Constancy

- Manipulation

- Randomization

- Pilot study

Pilot Study

A mini-version of the whole study beforehand to see how well it will work out

2 Main Types of Validity

- Internal

- External

Internal Validity

The degree to which the experimental treatment, not an uncontrolled condition, resulted in the observed effects

Threats to Internal Validity

- History

- Maturation

- Testing

- Instrumentation

- Mortality/Attrition

- Selection bias

History Threat

- Another specific concurrent event, apart from the independent variable, that may affect the dependent variable, either inside or outside the experimental setting

- Ex.) Effect of breastfeeding ads on research that is studying the effects of a breastfeeding education

program on the length of time of breastfeeding

Maturation Effects

- The developmental, biological, or psychological processes that operate within an individual as a function of time

- Ex.) Effect of growing older/wiser on research that is studying the effects of a specific teaching method

Testing Effects

- The effect on the participant's posttest score as the result of having taken a pretest

- Ex.) Individuals generally score higher when

they take a test a second time regardless of the

treatment

Instrumentation Threats

- Changes in the variables or observational techniques that may account for changes in the obtained measurement

- Ex.) An uncalibrated thermometer

Mortality/Attrition

The loss of study participants from the first data-collection point (pretest) to the second data-collection point (posttest)

Selection Bias

- The threat to internal validity that arises when pretreatment differences exist between the experimental group and the control group

- Ex.) Smoking cessation program is offered to all smokers; chances are, only the motivated ones will participate

External Validity

The extent to which the research findings can be generalized to real-world settings

Threats to External Validity

- Selection effects

- Reactive effects

- Measurement effects

Selection Effects

- A threat to external validity that is concerned with the generalizability of the results to other populations

- Important to set inclusion and exclusion boundaries

Reactive Effects (Hawthorne Effect)

- A threat to external validity that is concerned with the participants behaving in a certain way with the investigator not because of the study procedures but merely as an independent response to being studied

- Ex.) Students studying harder because they are being observed

Measurement Effects

A threat to external validity that is concerned with how an administration of a pretest in a study affects the generalizability of the findings to other populations; similar to testing effects

Methodology

A system of methods used in a particular area of study or activity (Ex. Quantitative, qualitative, Indigenous, and mixed methods)

Design

An overall plan for obtaining the information needed to address a research problem or issue (Ex. Randomized control trial, psychometric, phenomenology, non-experimental, etc.)

3 Main Types of Quantitative Designs

- Experimental

- Quasi-experimental

- Non-experimental

Experimental Design

- A research design in which their is randomization of participants and control and manipulation of variable

- Has the highest level of control

- Commonly called randomized clinical trials (RTC)

Advantages of Experimental Design

- Most appropriate for testing cause-and-effect

- Provides highest level of evidence for single studies

Disadvantages of Experimental Design

- Participant mortality, especially control group

- Difficult logistics in field setting

- Hawthorne effect

- Not all research questions can receive appropriate manipulation and randomization

3 Types of Experimental Designs

- True-experimental

- Solomon four-group

- After-only

True-Experimental Design (Classic Experiment)

Type of experimental design with a pre-test, post-test experimental and control group

Solomon Four-Group Design

- Type of experimental design with a pre-test/post-test experimental and control group, and an experimental and control group with no pre-test/pos-test

- Rules out testing effects

- Needs lots of people

After-Only Design

- Type of experimental design with an experimental and control group with no pre-test/post-test

- Only used if a base line is not needed

Key Points in an RCT

- Population

- Randomization

- Blinding

- Preciseness of results

Randomization

- The assignment of participants to either the experimental or the control group on a purely random basis (Ex. Flipping a coin)

- The more equal the groups are in size, the better

Blinding

A technique used in experimental and quasiexperimental research in which the participants are not aware of whether they are receiving the intervention

Double-Blinding

Level of blinding in which both the subjects and investigators are blinded

Triple-Blinding

Level of blinding where subjects, investigators, and managers of the study are blinded

Quasiexperimental Design

- A research design in which the researcher initiates an experimental treatment, but some characteristic of a true experiment is lacking (mainly randomization)

Advantages of Quasiexperimental Design

- Practical and more feasible, especially in clinical settings

- Some generalizability

Dissadvantages of Quasiexperimental Design

- Difficult to make cause-and-effect statements

- Not truly randomized

4 Main Types of Quasiexperimental Design

- Nonequivalent control group

- After-only nonequivalent control group

- One-group pretest post-test

- Time series

Nonequivalent Control Group Design

Type of quasiexperimental design with a pre-test/post-test experimental and control group, but no randomization

After-Only Nonequivalent Control Group Design

Type of quasiexperimental design with an experimental and control group with no pre-test/post-test and randomization

One-Group Pretest Post-Test Design

Type of quasiexperimental design with an experimental group with a pre-test/post-test, but no randomization

Time Series Design

Type of experimental design with an experimental group with two pre-test/post-test, but no randomization

Non-Experimental Design

- Type of research design that examines events, people, and situations as they naturally occur

- Constructs a picture of a phenomenon at one point or over a period of time

Advantages of Nonexperimental Design

- Important to develop knowledge base on a phenomenon of interest

- Useful in forecasting or making predictions

- Important when randomization, control, and manipulation are not appropriate/possible

- Useful in testing theoretical models of how variables work together

Disadvantages of Nonexperimental Design

Diffulty in explaining cause-and-effect relationships

5 Main Types of Non-Experimental Designs

- Survey research

- Relationship/difference studies

- Psychometric research

- Epidemiological

- Secondary analysis

Survey Research

- Type of non-experimental design that involves collecting detailed descriptions of existing variables through questionnaires/interviews

- Only gathers information and doesn't try to determine causation

Advantages of Survey Research

- Large amounts of information is gathered in a more economical manner

- Surprising accuracy of a small # of participants representing the views of the population

Disadvantages of Survey Research

- Information gathered tends to be superficial

- Breadth rather that depth of information is emphasized

- Requires a great deal of expertise in various research areas

- Can be time consuming and costly

2 Types of Relationship/Difference Studies

- Correlational

- Developmental

Correlational Studies

Type of relationship/difference study in which the investigator examines a relationship between two or more variables (not causation)

Advantages of Correlational Studies

- Flexibility

- Collects large amounts of data to examine the relationship

- Has otential for clinical application

- Foundation for future studies

- Explores relationship between variables that can't be manipulated

Disadvantages of Correlational Studies

- Inability to manipulate variables of interest

- No randomization in the sampling procedures

- Can't determine a causal relationship

3 Main Types of Developmental Studies

- Cross-sectional

- Longtitudinal/prospective

- Retrospective

Cross-Sectional Study

Type of developmental study where researchers examine data

at one time to explore relationships between variables

Longitudinal/Prospective Study

Type of developmental study where researchers examine data from the same group at different times

Advantages of Longitudinal Studies

- Each participant acts as their own control

- Increased depth of responses

- Early trends can be analyzed

- Higher level of control than retrospective studies

Disadvantages of Longitudinal Studies

- Data collection can take a long time, requiring a lot of resources

- Testing effects may be a threat

- Mortality is a significant threat

Retrospective Study (Ex Post Facto)

Type of developmental study where the researcher attempts to link current events to the past

Advantages of Retrospective Studies

- Similar to the correlational design

- Higher level of control than correlational

- More time efficient

Disadvantages of Retrospective Studies

- Unable to prove causation

- Alternative hypothesis may be the reason for the documented relationship

- Difficulty of finding similar group members

- Challenging to find clear data because memory is not perfect

Psychometric (Methodological) Research

Type of non-experimental design involve in the development and evaluation of data collection instruments, scales, and techniques

Epidemiological Studies

Type of non-experimental design that involves examining the factors affecting the health and illness of populations

Prevalence

The # of people affected

Incidence

The # of cases occurring in a particular period of time

Secondary Analysis

A form of research in which the previously collected and

analyzed data from one study are reanalyzed for a

secondary purpose

Systematic Reviews

A summation and assessment of research studies found in the literature based on a clearly focused question that uses systematic and explicit methods to identify, select, critically appraise, and analyze relevant data from the selected studies to summarize the findings in a focused area

Meta-Analysis

A type of systematic review that combines the statistical findings of two or more separate studies and analyzes them

Key Characteristics of a Systematic Review

- A clear, pre-defined research question

- A transparent, reproducible methodology

- A systematic search that attempts to identify all studies that would meet the criteria

- An assessment of the validity of the findings of the included studies

- A systematic presentation and synthesis of the characteristics and findings of the included studies

Advantages of Meta-Analysis

- Better estimate of treatment

- More complete evidence

- Produces a pooled or summary effect

Disadvantages of Meta-Analysis

Needs multiple studies that are somewhat similar

Sampling

The process of selecting representative units of a population for study in a research investigation

Inclusion Criteria

- Characteristics that prospective subjects must have to be in the study (Ex. You need to be a woman to participate in a study related to pregnancy)

- Needs a balance between narrowness and broadness

If the inclusion criteria is too narrow . . .

Results will not be generalizable

If the inclusion criteria is too broad . . .

- It will reduce the effect size

- There will be issues with compliance

Exclusion Criteria (Delimitations)

Characteristics that have clear contraindications of participating in a study (Ex. You can't participate in a milk study if your lactose intolerant)

Sample

A set of elements that make up the population

Element

The most basic unit about which information is

collected (Ex. Individuals, places, objects)

2 Types of Sampling

- Probability

- Nonprobability

Nonprobability Sampling

Type of sampling in which elements are chosen through non-random methods

3 Types of Nonprobability Sampling

- Convenience

- Quota

- Purposive

Convenience Sampling

Type of nonprobability sampling that uses the most readily available population (Ex. The first 25 people to volunteer)

Quota Sampling

Type of nonprobability sampling in which knowledge about the

population of interest is used to ensure some

representativeness about the sample; adds categorization

Probability Sampling

Type of sampling that uses randomization to assign elements

3 Types of Probability Sampling

- Simple random

- Stratified random

- Systematic

Simple Random Sampling

Type of probability sampling in which all units within a population are identified, and units are randomly selected (Ex. Table with random numbers associated with each units)

Stratified Random Sampling

Type of probability that is the same as quota sampling, but selection is random

Systematic Sampling

Type of probability sampling that

involves the selection of every “kth” case drawn

from a population list at fixed intervals (Ex. Picking every 10th member from the lists of volunteers)

3 Main Requirements of Data Collection

- Objective

- Systematic

- Consistent

4 Main Types of Quantitative Data Collection Methods

- Physiological/biological measurements (Ex. Vital signs)

- Observational methods

- Interviews/questionnaires

- Records or available data

4 Conditions of Scientific Observations

- Observations are consistent with the study's objectives and theoretical framework

- Standardized and systematic plan for observation and recording of data

- Training of data collectors

- All observations are checked and controlled

4 Types of Scientific Observation Methods

- Concealment with intervention

- Concealment without intervention

- No concealment without intervention

- No concealment with intervention

Example of Concealment with Intervention

A researcher secretly observes hand hygiene compliance in a hospital by installing hidden cameras and then implements an educational program to improve compliance

Example of Concealment without Intervention

A researcher discreetly observes customer shopping behaviors in a supermarket through security cameras without introducing any changes to the store layout or promotions

Example of No Concealment Without Intervention

A researcher openly observes classroom interactions by sitting at the back of the room and taking notes without influencing the lesson or student behavior

Example of No Concealment with Intervention

A researcher openly conducts a study on exercise habits by leading a fitness program in a community center while observing participants' engagement and progress

Validity

Refers to whether a measurement instrument accurately measures what it is intended to measure

4 Main Types of Validity

- Construct

- Content

- Face

- Criterion-related

Construct Validity

The degree to which a test measures the concept that it's intended to measure (Ex. Confirming that IQ test scores correlate with problem-solving ability and academic performance but not with unrelated traits like height)

Content Validity

The degree to which a test fully represents what it aims to measure (Ex. Making sure that the final exam includes questions that comprehensively cover all the key topics taught in the course, rather than focusing on just a few sections)