Chapter 1 - Stats Terms

Statistics

• set of mathematical procedures for organizing, summarizing and interpreting information

  • standardized techniques to interpret data and draw conclusions (so that the results are consistent across the world)

  • in a study, you collect data, and then use statistics to understand patterns in data

Population

• the group of people that you want to study

  • understanding mental health in first year students — population is first-year university students in Canada

• usually not possible to study the entire population so we use a smaller sample

Sample

• a set of people from the population that you test in the study

• intended to “represent” and “generalize” to the population

  • first year university students at MtA

A Variable

• something that can change/have different values for different people

  • ex; age, mental health, gender

Raw Score (Datum)

• a single measurement from a participant

Data/Data Set

• multiple measurements that are organized together in a file

Two Types of Statisitcs

1. Descriptive

2. Inferential

Descriptive Statistics

• procedures to summarize, organize and simplify data

• allows you to describe what’s going on (describe patterns in the data) but can’t draw any real conclusions from this information

Inferential Statistics

• techniques to study our sample data and make generalizations about the population

• remember what sample vs. population means?

• issue to be aware of — sampling error

  • systematic differences/effects vs. errors/noise

Sampling Error

• natural discrepancy between your sample data and the full population

Research Process

• develop a research question

• develop a hypothesis

• choose your sample, variables, and research design

• collect your data

• analyze your data

Research Design

• often, examining a relationship between variables

  • correlational method

  • experimental methods and (rarely) non experimental methods

Correlational Method

• measure 2 variables as they naturally occur in the world and see if there is a relationship between them

  • show relationship but cannot say that one caused the other (third variable problem)

Experimental Method

• manipulate one variable and see how it affects the second variable

Experimental Method — Requirements

• manipulate one variable

  • random assignment or matching (read that men enjoy these more so make sure that there are equal amounts of men in each condition)

• control for other variables so that we can systematically say that the results are based on the manipulation and not another factor

  • participant variables

  • environmental variables

Non-Experimental Method

• the same as the experimental method but you can’t randomly assign people to groups (studying age effects, gender effects…)

  • often called a quasi-independent variable

Operational Definition

• describes how your construct is measured and what it means

  • e.g., if you want to measure “hunger”

    • # of hours since last eating

    • rate on a scale from 1-7 (1 = not hungry, 7 = very hungry)

  • e.g., if you want to measure “intelligence”

    • score on an IQ test

• Puts your construct in tangible terms

“This proves…”

• NEVER SAY THIS

  • means its a fact no matter what — but this is not true in psychology

    • results can change depending on different people

• instead say “this suggests…”, “this demonstrates…”

Discrete Variables

• numbers that are finite, can be counted, and will be whole numbers

  • e.g., number of children, number of students in this room, # of dates someone has been on

Categorical Variables

• numbers that are used to represent distinct categories

• the numbers may not have a logical order

  • ex; experimental condition = 1, control condition = 2

  • yes = 1, no = 2

  • married = 1, single = 2

Continuous Variables

• numbers that have an infinite number of possible values and can be divided into an infinite number of fractional parts

• numbers can range by small amounts and rare to have people with the exact same number

  • e.g., time, height, weight, grades

*whenever you can choose the degree of precision or the number of categories for measuring a variable, it is continuous

Scales of Measurement

• nominal

• ordinal

• interval

• ratio

Nominal

• assigning numerical values to “name” categories

• the category values do not hold meaning

  • e.g., teacher = 1, lawyer = 2, therapist = 3

Ordinal

• assigning numerical values to categories in an ordered sequence

  • e.g., gold, silver, and bronze in a race = 1, 2, 3

    • the 1 being different from the 2 is a meaningful difference (finished faster)

• but…doesn’t tell you the difference between the categories

  • first and second were within a second but third was a minute later

Interval and Ratio

• assigning numerical values in an ordered sequence AND these must be of equal intervals

  • e.g., measurements in seconds

• allows us to see order AND direction/difference between categories

Interval

• zero is arbitrary and doesn’t indicate an actual true “zero”

  • e.g., 0 degrees does not mean there is no temperature

*rare doesn’t come up as often

Ratio

• the zero is meaningful as an absolute “zero”

  • e.g., a gas tank with 0 gas = empty; absence of gas

• allows us to describe things in terms of ratios

  • e.g., gas tank with 10 gallons has twice as much as a tank with 5 gallons

• more common

  • other examples: height, weight, reaction time, # of errors on a test