Page 15-16: Importance of Research and Statistics

Importance of Research and Statistics

• Research is fundamental in evidence-based clinical practice.

• Statistics supports quantitative research; qualitative research is also valued.

Research Time Lag

• Average 17-year gap between discovery and practice.

• Highlights the importance of knowledge generation in nursing.

What is Data?

• Data conveys information collected to answer research questions.

• Comes in various forms such as surveys and experiments.

What are Statistics?

• Involves collection, analysis, interpretation, and presentation of numerical data.

• Statistical tests compare data expectations vs. actual results.

Levels of Measurement

• Dictates the statistical tests used.

Nominal Level

• Nominal variables are categories (e.g., blood type, eye color).

Ordinal Level

• Ordinal variables imply an order among categories (e.g., happiness scale).

Interval Level

• Known distances between scores but no true zero (e.g., temperature).

Ratio Level

• Known distances and a true zero (e.g., weight, income).

Continuous vs. Categorical Variables

• Nominal/ordinal are categorical; interval/ratio are continuous.

Primary vs. Secondary Data

• Primary data: Collected directly by researchers.

• Secondary data: Existing data reused for new research.

Population vs. Sample

• Population: Large dataset hard to access; samples provide estimates.

Types of Papers in Journals

• Categories: Primary articles, secondary articles, special articles.

Research Question

• Addresses a problem, should be precise and answerable.

• May involve a hypothesis in quantitative research.

Operationalizing a Variable

• Concepts: Ideas defined formally for measurable variables.

• Examples include: Pain measured by a 1-10 rating scale.

Descriptive Statistics

• Summarizes data meaningfully; provides insight into average.

Measures of Central Tendency

• Mean: Arithmetic average.

• Median: Middle value of sorted data.

• Mode: Most frequently occurring value.

Measures of Dispersion

• Range: Difference between max and min values.

• Variance: Average squared differences from the mean.

• Standard Deviation: Square root of variance.

Interpretation of Standard Deviation

• Small SD: Data points close to the mean.

• Large SD: Data widely spread from the mean.

Inferential Research

• Makes inferences about a population from a sample.

• Questions: "Is X associated with Y?" or "Does X cause Y?"

Compare Descriptive vs. Inferential Research

Descriptive Statistics

• Summarize and describe data; provide an overview of dataset characteristics.

Inferential Statistics

• Make predictions and test relationships; involves hypothesis testing.

Variables

Independent Variable

• The factor manipulated by researchers (e.g., medication type).

Dependent Variable

• The outcome measured (e.g., reduction in patient pain).

Understanding Probability

• Definition: Likelihood of an event occurring (0% to 100%).

Normal Distribution

• Characteristics: Mean, median, and mode are equal.

• Symmetrical and often appears in nature.

Standard Deviation and Probability

• Total probability = 100%; understanding of probability by referencing standard deviation.

Sample Statistics

• Sample Mean Height: 165 cm; SD: 5 cm.

• Key percentages for height ranges.

Standard Error vs. Standard Deviation

• SD: Measures variability of observations.

• Standard Error: Indicates closeness of sample mean to population mean.

Expanding to Population Estimates

• Estimate true population mean using sample statistics.

Confidence Intervals

• Confidence Level: 68% to 95% likelihood.

Introduction to Hypothesis Testing

• Definition: Inferential research with implications on populations.

Statistical Significance

• Definition: Probability that differences are not due to chance.

Understanding P-Value

• Indicates probability of observed outcome given null hypothesis is true.

Null Hypothesis

• Represents no association.

Alternative Hypothesis

• Represents existence of an association.

Understanding Probability Samples

• Known probability of selection; includes simple random, systematic, stratified, cluster.

Non-Probability Samples

• Unknown probability of selection; includes convenience sampling and quota sampling.

Reasons for Non-Probability Sampling

• Acceptable when results are not intended to extrapolate to larger populations.

Statistical Tests

• Factors include: Number of groups, sample size, and data distribution.

Parametric vs. Non-parametric Tests

• Parametric: Assume specific distribution.

• Non-parametric: Fewer assumptions; used for non-normally distributed data.

When to Use Each Test

• Use parametric tests with large samples under normality.

• Use non-parametric tests with small samples or skewed data.

Common Statistical Tests

• Chi-square, T-test, correlation, regression.

Chi-square Test

• Null Hypothesis: No relationship between variables.

Correlation Tests

• Pearson's r: Measures linear relationships.

T-test

• Null Hypothesis: No significant difference between group means.

ANOVA

• Null Hypothesis: No significant differences among group means.

Interpretation of Statistical Tests

P-Values and Confidence Intervals

• Understanding statistical significance in correlation, association, and regression analyses.

Examples of Statistical Applications

• Independent t-test: Risk factors in patients.

• Paired t-test: Heart rate measurement comparisons.

• ANOVA: Analysis of Internet user types.

Odds Ratio Interpretation

• OR = 1: No association; OR > 1: Positive association; OR < 1: Negative association.