1.1 -1.2

Foundations of Statistics and Data

• Statistics is defined as the formal study of five key processes: the collection, organization, summarization, analysis, and interpretation of data.

• The primary purpose of statistics is to utilize data to answer specific questions when uncertainty is present.

• Through the application of statistical methods, researchers can transition from mere observations to making evidence-based decisions.

• An example application of statistics is estimating whether a group of students prefers online textbooks or printed textbooks.

Data Basics and Raw Information

• Data consists of specific pieces of information gathered from various sources, including people, objects, events, or processes.

• Common examples of data include:

  • Exam scores.

  • Commute times.

  • Blood types.

  • Product ratings.

  • Rainfall amounts.

• Raw data usually requires systematic organization before any underlying patterns or trends become visible to the researcher.

Individuals and Variables

• Individuals represent the specific units being studied in a research project.

• Variables are the specific characteristics that are measured or observed from those individuals.

• In a study focusing on students:

  • The individuals are the students.

  • The variables may includes attributes such as age, GPA, major, and study time.

• In a study focusing on cars:

  • The individuals are the cars.

  • The variables may include price, mileage, fuel type, and model year.

Classification of Variables: Quantitative and Qualitative

• Quantitative Variables: These are numerical in nature. Arithmetic operations, such as calculating an average or a sum, are meaningful for these variables. Examples include:

  • Height.

  • Income.

  • Number of siblings.

  • Test scores.

• Qualitative Variables: Also known as categorical variables, these describe a specific category or group. Examples include:

  • Eye color.

  • College major.

  • Vehicle type.

  • Membership status.

Discrete and Continuous Variables

• Quantitative variables are further divided into discrete and continuous categories:

  • Discrete Variables: These have countable values, often involving whole numbers. Examples include the number of children, the number of emails received, the number of books owned, or the number of defective items.

  • Continuous Variables: These can take on any value within a given interval or range. They are typically measurements. Examples include height, weight, time, temperature, and distance.

Population, Sample, and Census

• Population: The entire group of interest that a researcher wishes to study (e.g., all students at a university).

• Sample: A smaller, selected group chosen from the population for actual measurement or observation (e.g., $300$ selected students).

• Census: A method of data collection where information is gathered from every single individual within the entire population. For example, recording the exam scores for every student in one specific class qualifies as a census of that class.

Parameters and Statistics

• Parameter: A numerical measure that describes a characteristic of an entire population. This value is often unknown and is what researchers seek to estimate. An example is the mean GPA of all students at a college.

• Statistic: A numerical measure that describes a characteristic of a sample. This is a calculated value used to estimate the population parameter. An example is the mean GPA of $150$ selected students.

Descriptive and Inferential Statistics

• Descriptive Statistics: This branch focuses on summarizing and describing the data that has been collected. Tools used in descriptive statistics include:

  • Tables.

  • Graphs.

  • Measures of central tendency (mean, median).

  • Measures of dispersion (range, standard deviation).

• Inferential Statistics: This branch involves using sample data to draw conclusions or make predictions about a larger population. This includes estimating population means or conducting tests to determine if a specific treatment is effective.

Levels of Measurement

Data can be classified into four nested levels of measurement, determined by whether the data can be ordered, if differences are meaningful, and if a true zero exists:

• Nominal: These are categories with no natural or inherent order. Example: Blood type.

• Ordinal: These are categories that can be placed in a specific order, but the differences between the categories are either not equal or cannot be determined. Example: Class rank.

• Interval: These are ordered numerical data points where the differences between values are meaningful, but there is no "true zero" point (a value of zero does not represent the total absence of the characteristic). Example: Temperature in Celsius ($^{\circ}C$).

• Ratio: These are ordered numerical data points where the differences are meaningful and there is a "true zero" point (where zero represents the complete absence of the characteristic). Ratios are also meaningful at this level. Example: Weight and Monthly Rent.

Practical Application: Grocery Store Study

• In a scenario where a researcher studies $150$ grocery-store customers and records the amount spent, the payment method, and the number of items purchased:

  • The individuals are the $150$ grocery-store customers.

  • The variables are amount spent, payment method, and number of items purchased.

  • Classification of variables:

    • Amount spent is a quantitative variable.

    • Payment method is a qualitative variable.

    • Number of items purchased is a quantitative variable.

Practical Application: University Parking Survey

• In a scenario where a university has $12,000$ students and a researcher surveys $300$ students regarding campus parking, calculating the sample proportion of dissatisfied students:

  • The population is all $12,000$ students at the university.

  • The sample is the $300$ students who were surveyed.

  • The calculated proportion is a statistic because it is computed from the sample data rather than the entire population.

Practical Application: Identifying Levels of Measurement

• Blood Type: Classified as Nominal because it is a category without a natural ranking or order.

• Class Rank: Classified as Ordinal because the ranks represent a specific order.

• Temperature in Celsius ($^{\circ}C$): Classified as Interval because the differences between degrees are meaningful, but $0^{\circ}C$ does not mean an absence of temperature (no true zero).

• Number of Books Owned: Classified as Ratio because it is a count where zero represents the absence of books.

• Monthly Rent: Classified as Ratio because a value of zero means no rent is being paid, and the ratio of one rent to another is meaningful.

Practical Application: Descriptive vs. Inferential Analysis

• In a scenario where a polling organization surveys $1,000$ voters and finds that $54\%$ support a proposed law, and then uses this to estimate support among all voters:

  • The descriptive component is the specific result: $54\%$ of the surveyed voters support the law.

  • The inferential component is the act of using that $54\%$ sample result to estimate the level of support among the entire population of voters.

  • The role of the sample is to provide evidence and a basis for making conclusions about the larger population.