1b

HP 340Lg: Introduction to Health Behavior Statistical Methods

Introduction to Statistics:
- Read Chapter 1, pp. 1-16 (Importance of Statistics)
- Read Chapter 2, pp. 17-30 (Types of Data)

Definition of Statistics:
- Statistics is a branch of mathematics dealing with the collection, analysis, interpretation, presentation, and organization of data.
Data Collection Importance:
- The insights gained from data are heavily dependent on how it was collected; biases in data collection can lead to erroneous conclusions.
Types of Studies:
- Various studies can be employed such as anecdotal evidence, observational studies, randomized controlled trials, and cohort studies, each with its strengths and weaknesses for drawing valid conclusions.

Types of Data:
- Quantitative Data:
  - Involves measurable characteristics and uses numbers, allowing for statistical analysis (e.g., temperature, weight, age).
- Qualitative Data:
  - Consists of non-numeric data; categorized into groups for analysis (e.g., eye color, types of diseases).

Definition:
- Continuous data can take any value within a range and have fractional values (e.g., height = 65.125 inches).
Statistical Measures:
- Mean, median, mode, and standard deviation can be calculated, providing insights about central tendency and data dispersion.
Visualization Techniques:
- Histograms are used to show the distribution of continuous data, while scatter plots visualize relationships between two continuous variables, highlighting trends and correlations.

Definition:
- Discrete data is numeric but can only take whole number values (e.g., count of cars per household).
Statistical Measures:
- Calculating the mean may not apply as fractional values do not make logical sense (e.g., saying there are 2.11 cars).
Visualization:
- Bar charts effectively represent frequency counts or values associated with discrete categories, helping to compare different groups easily.

Categories:
- Qualitative data can be organized into ordered (ordinal) or unordered (nominal) categories.
- An example of ordinal data may include a ranking of health statuses (e.g., poor, fair, good), while nominal data might categorize participants based on blood type.
Visualization Techniques:
- Categorical data can be visualized using pie charts to highlight proportionate distributions, with ordinal data often represented in a way that shows relative order but not necessarily equal intervals.

Populations vs. Samples:
- Understanding the distinction between a population (the entire group of interest) and a sample (a subset of the population) is crucial for statistical inference.
Statistical Inference:
- This process is essential when full populations cannot be studied; it allows researchers to make predictions or generalizations based on samples.

Example Study:
- A selected LA County sample was used to study COVID-19 impacts; the method involved defining the population and appropriate sampling techniques.
Define Population and Sample:
- For example, a sample of 3000 participants was used to infer health behaviors for an entire population of 10 million residents, illustrating principles of statistical inference.
Population and Sample Understanding:
- The study focused on antibody detection across various populations, evaluating the methodology employed and the findings derived from this data.

Overview of Statistics:
- An introduction to the fundamental aspects of statistics including its definition, polling methodology, sampling techniques, margin of error, and the various types of variables (both numeric/discrete & categorical/nominal) as well as the usage of SPSS for statistical analysis.
Statistics Defined:
- Learning how to draw conclusions from data through mathematical methods is essential for effective analyses.
Statistical Methods Covered:
- The course will cover 12 different statistical methods, with a focus on their practical applications without delving too deeply into complex mathematical derivations.
Uses of Statistics:
- An illustrative case discussed is the relevance of polling data (e.g., public opinion on political candidates), emphasizing the importance of understanding margins of error and sample representation in understanding results.

Representative Samples:
- Crucial for drawing valid conclusions, representative sampling involves techniques such as:
- Simple Random Sample: Every individual has an equal chance of selection.
- Stratified Random Sample: Population is segmented into different strata and samples are drawn from each.
- Additional sampling methods exist to minimize biases and enhance representativeness.

Relevance in Careers:
- Essential for various professions including scientists, public health professionals, and journalists, as these roles often require the interpretation and communication of statistical data.
Combatting Misinformation:
- A thorough understanding of statistics permits clearer explanations of scientific findings and can help address and reduce misinformation circulating in public discourse.

Categorical Data:
- Falls into exclusive categories; may be further divided into:
- Nominal: No natural order among categories (e.g., types of fruit).
- Ordinal: There exists a natural order among categories (e.g., stages of disease).
Numeric Data:
- Can consist of both whole (discrete) and real (continuous) numbers, allowing for various types of statistical analysis.
- Discrete: Countable data, usually whole numbers (e.g., number of patients).
- Continuous: Can take any value in a specified range (e.g., blood pressure readings).

Graphing Continuous Variables:
Histograms:
- Used to reveal the shape of data distributions, including central tendency and dispersion, by grouping data into bins.
Graphing Categorical Variables:
- Pie Charts:
  - Effective for showing proportions of different categories.
- Bar Charts:
  - Useful for categorical data comparisons with clear distinctions between group frequencies/values.
Distinguishing Charts:
- Bar Charts are used for categorical data representation while Histograms provide grouped continuous data representation, important for understanding the nature of the data.

SPSS:
- The focus will be on importing datasets and performing data cleaning in the initial stages.
To-Do List:
- Complete assigned readings and homework prior to the next class to stay abreast of core concepts and practical applications of statistics in health behavior research.