HINT 110 Final
Inferential Research
Definition: Inferential research involves the use of inferential statistics to make inferences about a population based on a sample drawn from that population.
Key Questions:
Is X associated with Y?
Does X cause Y?
Descriptive vs. Inferential Research
Descriptive Research:
Purpose: To summarize and describe data, providing a clear overview of characteristics within a dataset.
Presentation: Data is presented using measures of central tendency, variability, or visual graphs.
Functions:
Generates initial understanding of a topic
Formulates hypotheses
Presents data in an understandable manner.
Inferential Research:
Purpose: To make predictions and draw conclusions about populations based on sample data.
Functions:
Hypothesis testing to make informed decisions.
Assess significance of relationships/differences.
Draws conclusions that can be generalized to a broader population
Example: Assessing if a research variable works across a population by inferring from sample results.
Research Questions (Descriptive)
Example Question:
What proportion of tubal sterilizations in Canada are performed via tubal ligation compared to opportunistic salpingectomy? (Kaur P et al., 2023).
Reference: Journal of Obstetrics and Gynaecology Canada.
Another Example Question:
What proportion of households in Europe experience catastrophic health spending? (Thomson S et al., 2024).
Reference: The Lancet Regional Health - Europe.
Type: Descriptive research does not involve inference/hypothesis testing; it observes existing occurrences.
Independent and Dependent Variables
Independent Variable:
Definition: The factor manipulated or controlled in an experiment. This variable is presumed to influence or affect the dependent variable.
Example (Experimental Research): Type and dosage of medication.
Example (Observational Research): Effect of smoking on lung health.
Dependent Variable:
Definition: The outcome or response measured as a result of changes in the independent variable.
Example: Reduction in patient pain levels post-medication.
Characteristics of Research Types
Descriptive Research: No independent variable; simply describes characteristics.
Inferential Research: Involves independent and dependent variables, investigating cause-and-effect relationships.
Inferential Research Questions
Key Example:
Arundel C et al. (2025): Negative pressure wound therapy vs. usual care in surgical wound healing.
Independent Variable: Type of care received.
Dependent Variable: Healing time.
Note: Inferential questions evaluate how well research findings can apply to wider populations.
Final Thoughts on Research Types
Research papers may present research questions both explicitly and implicitly.
Papers can contain both descriptive and inferential questions.
Inferential papers often focus on samples to extrapolate conclusions to larger populations.
Operationalizing a Variable
Concept Definition: Concepts (e.g., health, illness, anxiety) must be formally defined to enable their study.
Operational Definitions: These include methods of measurement.
Example: Pain as a concept can be measured using a 1-10 rating scale.
Socioeconomic status can be defined by yearly income.
Importance: Effective operational definitions help gauge how well the variable reflects the initial concept.
Descriptive Statistics
Purpose: Summarizes data meaningfully in health research.
Insight Offered: Provides average or central tendency.
Key Questions: Always ask, “What’s happening?”
Measures of Central Tendency
Common Measures:
Mean: Arithmetic average of all data points.
Median: Middle value when data is sorted.
Mode: Most frequently occurring value.
Usage: Frequently used for summarizing characteristics of variables (e.g., age, gender, diagnosis).
Examples of Central Tendency
Reference example of dataset from Kaggle (quantitative data).
Measures of Dispersion or Variability
Common Measures:
Range: Difference between max and min values.
Variance: Average of squared differences from the mean.
Standard Deviation: Square root of variance indicating data spread.
Implication: A small standard deviation indicates data points are closer to the mean (less variability), while a large standard deviation shows greater spread (more variability).
Statistical Tests and Analysis Framework
Fundamental Tests: Understand tests like t-tests and ANOVA, and recognize when to apply them.
Considerations: Factors to consider in statistical tests include the number of groups, group relationships, sample size, and data distribution (independent vs dependent).
Parametric vs. Non-Parametric Tests
Parametric Tests: Assume normal distribution and require specific statistical criteria. Sensitive to data conditions and sample size.
Non-Parametric Tests: Make minimal assumptions about data and are used when data does not fit parametric assumptions.
Examples include Wilcoxon signed-rank test and Mann-Whitney U test.
Bias in Research
Definition of Bias: Systematic errors or deviations from accurate results, can be conscious or unconscious.
Includes confirmation bias, leading to favoritism towards confirming existing beliefs while ignoring contradictory evidence.
Reasons for Bias: Oversight of good research methods, inadequate data, publication pressure, or experimenter biases.
Type I and Type II Errors
Type I Error: Rejecting the null hypothesis erroneously when it is true; leads to false positive results.
Type II Error: Failing to reject the null hypothesis when it is false; leads to false negative results.
Sampling and Confidence Intervals
Sampling Error: Differences between sample and population means, often addressed by 95% confidence intervals to estimate true parameters.
Sampling Bias: Systematic error regarding sample selection failing to represent the population effectively.
Example: Only including certain demographic groups in a nursing study can skew results.
Confounder Bias
Definition: An association created by an unaccounted variable influencing both the exposure and the outcome.
Example: The correlation seen between red wine consumption and heart health can be confounded by socioeconomic status which also affects health choices.
Conclusion on Research Types
Different types of research address varying questions in public health and nursing.
The understanding of how variables interact is crucial to effectively analyze results and draw valid conclusions.
Bias checks and careful methodology are required to ensure robust findings.