NOTES - 1 Introduction to Statistics Outline •Populations and Samples •Sampling Error •Measurement and Scales •Experimental Method •Order of Operations Research Question: How many hours of sleep do CBU students get? Characteristics of populations and samples •Variable • Characteristic or condition that changes or has different values for different individuals Characteristics of populations and samples •Variable • Characteristic or condition that changes or has different values for different individuals •Data • Measurements or observations of a variable ◦ Single data point commonly called a score or raw score Population and Sample Population Sample Population • Set of all individuals of interest • Often large (but not always) Sample • Set of individuals selected from a population • Intended to represent the population Population and Sample All CBU students This class Question: How many hours of sleep do CBU students get? Population • All CBU students Sample • Students in this Quant I class Parameters and Statistics • Parameter: A value that describes a population. ◦ Derived from measurements of the individuals in the population • Statistic: A value that describes a sample • Derived from measurements of the individuals in the sample Descriptive Statistics •Descriptive statistics can tell us the characteristics of sample • What is the average score? • What is the range of observed values? (highest and lowest) • How are the scores spread out? •Unit 1 (Chapters 2-4) will focus on descriptive statistics Inferential Statistics •Inferential statistics allow us to use samples and make generalizations to the population • Is the observed outcome due to chance? Sampling Error •Sample is an approximation of the population ◦ Never exactly the same •Sampling error: the discrepancy (amount of error) between and sample statistic and the corresponding population parameter • Unavoidable • Usually random Sampling Error: Application Population: CBU Students 7 hrs Group 1 8 hrs Group 2 8 hrs How many hours of sleep do CBU students get per night? Group 3 4 hrs Sampling Error or an effect? 1. Is the difference due to sampling error (“due to chance”)? 2. Is the difference not “due to chance”? ◦ Usually explained by “an effect” Inferential statistics will help us determine this Quick Review So Far What is a value that describes a population? What is a value that describes a sample? Why are values for a sample not equal to the values of population? Outline •Populations and Samples •Sampling Error •Measurement and Scales •Experimental Method •Order of Operations Discrete and Continuous Variables •Discrete Variables • Have separate categories • No values exist between two neighboring categories • Examples: Brand, number of people, letter grade in class (A-F) Discrete and Continuous Variables •Continuous Variables • Infinite number of possible values between any two observed values • Every interval is divisible into an infinite number of equal parts ◦ E.g., 5 → 2.5 → 1.25 → .625 • Examples: Height, Weight, Number-lines Discrete and Continuous Variables •Discrete Variables • No values exist between “apple” and “orange” • Continuous Variable • An infinite number of values exist between 5 and 6 ◦ 5.2, 5.08, 5.0001, . . . Scales of Measurement •Nominal scale •Ordinal scale •Interval scale •Ratio scale Nominal scale • Categories with names • Does: Label data • Does NOT: do anything else • Examples: • Room numbers • Colors • Brand Ordinal scale • Categories in an ordered sequence • Does: Place data in a ranked order • Does NOT: Tell us anything about quantitative differences • Examples: • Olympic medals • French fry size (S/M/L) Interval scale • Ordered categories with equal intervals between categories • Does: Place data in order, with equivalent distances between the ordered categories • Does NOT: Have a meaningful zero ◦ A value of “0” does not indicate a true absence of the variable • Examples: • Fahrenheit/Celsius, IQ Ratio scale • Ordered categories with equal intervals between categories AND a “meaningful” or “true” zero point • A value of “0” indicates a true absence of that variable • Does: Basically an Interval scale where zero means something • Does NOT: This is as good as it gets for us • Examples: Money, Speed, Gas tank Scales of Measurement Overview Why do scales matter? •Certain statistical procedures are only appropriate for some types of data • Most of the tests we’ll be working are only for interval or ratio data Learning Check A study assesses the optimal size (number of other members) for study groups. The variable “Size of group” is … Learning Check – Answer A study assesses the optimal size (number of other members) for study groups. The variable “Size of group” is … Outline •Populations and Samples •Sampling Error •Measurement and Scales •Experimental Method •Order of Operations Three data structures 1.Correlational Method Relationships between variables/groups 2.Experimental Method Demonstrate cause and effect Experimental Method •Goal of Experimental Method • To demonstrate a cause-and-effect relationship between two variables •One variable is determined by the experimenter •The other variable is measured Independent/Dependent Variables •Independent Variable is the variable manipulated by the researcher •Dependent Variable is observed to assess the effect • Dependent because its value is thought to depend on the value of the independent variable Experimental Method •Experimental condition • Individuals do receive the experimental treatment •Control condition • Individuals do not receive the experimental treatment. • Purpose: to provide a baseline for comparison with the experimental condition Experimental Method: Example •Does drinking coffee cause you to study more? Experimental Method: Example •Does drinking coffee cause you to study more? • Change the level of “coffee” ◦ 0 oz ◦ 8 oz ◦ 16 oz Experimental Method: Example •Does drinking coffee cause you to study more? • Change the level of “coffee” ◦ 0 oz ◦ 8 oz ◦ 16 oz • Measure number of hours spent studying Statistical Notation Statistical Notation Notation Definition X Individual scores for a particular variable Y Individual scores for a particular variable N Number of scores in a population n Number of scores in a sample Σ Summation of a symbol or equation Order of Operations Please Excuse My Dear Silly Aunt Sally Summation Notation Example ΣX = ? ◦ “Sum of X” ◦ 1+1+1+1+1+1 = 6 ΣY = ? ◦ “Sum of Y” ◦ 1+2+3+1+2+3 = 12 X 1 1 1 1 1 1 6 Y 1 2 3 1 2 3 12 Why aren’t these equations equal? a) ΣY2 vs. b) (ΣY)2 ◦ a) Square each Y score, add up the squared scores ◦ b) Add the Y scores, square that sum Why aren’t these equations equal? a) ΣY2 vs. b) (ΣY)2 ◦ a) Square each Y score, add up the squared scores ◦ b) Add the Y scores, square that sum a) ΣY2 Y 1 2 3 1 2 3 Y2 12 22 32 12 22 32 Y2 1 4 9 1 4 9 28 Why aren’t these equations equal? ΣY2 (ΣY)2 a) vs. b) ◦ a) Square each Y score, add up the squared scores ◦ b) Add the Y scores, square that sum b) (ΣY)2 Y 1 2 3 1 2 3 12 2 = 144 Review Research Question: “Does giving cats more treats lead to less biting?” ◦ What is the population? ◦ What is the Independent variable? ◦ What is the Dependent variable? ◦ What type of variable is this (discrete or continuous)? ◦ What scale of measurement? (nominal, ordinal, interval, or ratio) Review What is ΣX + 5 ? What is Σ(X – 2) X 7 3 9 Review What is ΣX + 5 ? Sum of all X, then add 5 7+3+9 = 19 19+5 = 24 What is Σ(X – 2)? Subtract 2 from each X, then sum 7 - 2 = 5, 3 - 2 = 1, 9 - 2 = 7 5+1+7 = 13 X 7 3 9 Questions?