Module 1, Intro to Statistics & the Scientific Method Flashcards
Quantitative
the data is in numbers
What is Empirical Research?
empirical research: any activity in which data are collected from some area of experience and then conclusions are drawn from the data about the area of experience
- captivates both quantitative and qualitative research
- it means we collect data and make sense of it (interpret)
- through stats we make sense of that data - a tool we use to help make sense of quantitative research
Branch of Mathematics
- collection - quantitative data
- analysis (for example: going through procedures of doing a particular mathematical operation on that data to come up with an average)
- *interpretation - what do the numbers mean (something is just a number until it is interpreted)
- presentation of numerical data
The Scientific Method (5 steps)
- developing a research hypothesis to be tested (distinguishing between variables & types of hypotheses)
- collecting data (sample vs. population, levels of measurement & experimental vs non-experimental methods)
- analyzing data (descriptive vs inferential statistics)
- conclusions about research hypotheses (use of language)
- communicating findings
* loop back to research hypothesis to draw conclusions (loop 4 to 1)
What is a research hypothesis?
research hypothesis: a statement regarding an expected or predicted relationship between variables (they are very specific as they have to be testable)
Variable
variable: a property or characteristic that can take on different values
- variables are related to quantitative research as they take on different values (data will be in numbers)
- a variable needs to vary - cannot just take on one value (ex. class attendance (whether you attended or not) varies, not who attended today)
- measure is a key word: for instance, measure the degree to which someone is experiencing depressive symptoms (special case where you cannot directly measure something)
Construct
constructs is when you cannot directly measure a variable, they are theoretical (that means the definition of what it is can change)
‣ mental toughness - there
are many definitions of it
making it a construct
‣ constructs are a special
type of variable
How do we determine what the research hypothesis is?
research hypotheses can come from a variety of sources:
1. identifying a question or issue to be examined
2. *review and evaluating relevant theories and research (help guide us in selecting what variables we should use and how they connect)
How do we Distinguish Between Variables? (IV/DV)
independent variable (IV): the variable manipulated by the researcher
dependent variable (DV): the variable measured by the researcher
- want to know the effect of IV on the DV
example: the thing that is being manipulated is the number of grams of proteins (IV) (after a workout) - 20g, 40g, 60g (this variable is varying) - levels of an independent variable (same) are the different numbers
the number of groups you have is the same as the number of levels of the IV - 3 levels in this case
Directional Hypothesis
is going to state whether a group scores higher or lower on an outcome variable compared to another group and could indicate whether there is a positive or negative relationship
- mostly there are directional hypotheses that are used to see which group will do better
Non-Directional Hypothesis
would state that there is a relationship between the variables but do not know if there is a positive or negative relationship OR we do not know which group will score higher or lower on the outcome variable (just know there will be a difference)
Among both directional and non-directional hypothesis which is often utilized more?
directional
Research Question: Do youth soccer players who do not wear head protection experience a different rate of concussions than those who wear head protection?
(IV/DV/HOW MANY LEVELS/WHAT LEVEL OF MEASUREMENT)
DV = # of concussion (ratio level of measurement - true zero point)
IV = presence of head protection (variables have to vary) (2 levels of the IV - nominal/categorical)
- the presence of head protection will have an impact on the number of concussions (hypothesis causal relation) IV -> DV
Research Hypotheses: There will be a difference in the number of concussions between youth soccer players who wear head protection and those who do not
research
Q: IS IT DIRECTIONAL OR NON-DIRECTIONAL HYPOTHESIS?
non-directional (difference)
Research Hypotheses: Youth soccer players who wear head protection will have fewer concussions than those who do not wear head protection
Q: IS IT DIRECTIONAL OR NON-DIRECTIONAL HYPOTHESIS?
directional (fewer)
- are stating who is going to score higher or lower on the outcome variable
Research Hypotheses
- the relationship between variables written with words
◦ A related to B
◦ A causes B
Statistical Hypotheses
- are what we are actually going to test, are mathematical representations of what we are testing
- will not be in words rather mathematical symbols
Statistical - Null Hypothesis
- states there is NO relationship
- statistical hypothesis to be rejected
- says nothing will happen/no relationship
Statistical - Alternative Hypothesis
- often associated with research hypothesis, in that it states there will be a relationship or a difference
- mathematical symbols
- how you state them will change/differ depending on the test you will conduct or do (for both)
Research Hypotheses: There will be no difference in the number of concussions between youth soccer players who wear head protection and those who do not
Q: IS THIS NULL OR ALTERNATIVE HYPOTHESES?
null hypothesis (states there will b NO dfference)
Population
population: the total number of possible units or elements that could be included in a study
◦ theoretical population
◦ often times the unit is people
(other examples include team,
sport, animals, league,
university)
Sample
sample: a subset of the population used to represent the population
- if you are in the sample you are also in the population
- sample characteristics = population characteristics (sample characteristic is representative of population characteristics)
Measurement
concerned with the methods to provide descriptions of the degree (value) to which an individual (or place or thing) possesses a defined characteristic (property)
- the quality of data and measurement is very important so the interpretation is accurate
Why does the level of measurement matter? the type of scale effects:
- what we can do statistically with the data
- the mathematical operations that can be performed (the certain level of measurement has certain assumptions attached to it so we cannot do certain math operations on that scale sometimes)
- how we interpret data
- whether differences (between individuals or groups) are meaningful
- whether larger/smaller numbers are “better”?
What are the 4 Levels of Measurement used for variables?
- measurement is the cornerstone to quantitative research
levels of measurement:
1. nominal
2. ordinal
3. interval
4. ratio
Nominal (categorical)
values differ in category or type (a numerical value is used to denote a category)
- the number itself is not valuable where you can assign numbers to sports to categorize them for instance as the not indicate rank or order
- the scale only represents a category and does not represent rank
- independent variables are typically measured at a nominal level of measurement - typically based on categories - a control group and experimental are also considered a category
Levels of Measurement (nominal)
refers to the number of levels of measurement within a nominal variable
- amounts to the number of categories
◦ control groups are included in
levels of measurement
◦ level of measurement is not an
actual variable (for example,
football is a level of
measurement but variable is
the sport type)
Ordinal
values can be placed in order relative to other values
- think about ranking (we only know rank not how good or bad someone was in comparison to someone else)
Interval
values are equally placed along a numeric continuum - no absolute zero
- temperature for example, there is zero degrees but it does not denote that there is no temperature and thus no absolute zero
rating scales (1-5) - numeric continuum -> this type of scale can be seen as ordinal or interval - reverse coded - the value becomes the opposite where 5 becomes 1 and 1 becomes 5 (pay attention to how things are worded)
Ratio
values are equally spaced on a numeric continuum - true zero point
- what is measured is the distance you can run in 12 minutes (the distance (m) is the scale) - the scale itself has a true zero - starting at zero
Likert Type Scales
the measurement scale is arbitrary and then you have different response options
- anchors on other ends like strongly agree and strongly disagree (usually disagreement scales)
- if you have likert type scale that has 4 or less response options = ordinal level of measurement
- if you have a likert type scale that has 5 or more response options = interval level of measurement
◦ this rule of thumb is only for
likert type scales
Discrete Variables
- no underlying continuum exists
- measure classifies items into non-overlapping categories
◦ nominal/categorical and
ordinal scales
Continuous Variables)
- underlaying continuum
- eg. aggression, reaction time, mental toughness
- interval and ratio level of measurement
there are different stats that we would use for continuous or discrete particularly when looking at outcome variable or IV
Descriptive Statistics
- organize, summarize and describe the data that has been collected
◦ collecting data from the
sample usually
◦ measures of central
tendencies - the average or
mean, median and mode
(most frequently occurring
data)
◦ measures of variability -
standard deviation, range and
the variance
Inferential Statistics
- test hypotheses and draw conclusions about the data collected from the sample
- inferences from samples to populations (interested in making inferences about a population)
◦ underpinning assumption mis
that you have a representative
sample of whats actual going
on in the population - mean arm hang of this sample could be used to represent the mean arm hang of all female Canadians aged 20-29
Conclusions about Research Hypotheses - Use of Language
- ask whether or not the results support the research hypothesis
◦ there is an important
distinction between support
and prove
◦ prove is based on certainty but
we want to look at probability
◦ inferential statistics is based
on probability
Communicating Findings
- communicating the results and interpretation of the results is important within the field
◦ publishing academic journals,
graph, infographic (visuals),
present at conferences, social
media
Experimental Research Methods
are methods designed to test causal relationships between variables—more specifically, whether changes in independent variables produce or cause changes in dependent variables. To make inferences about cause-effect relationships, researchers conducting an experiment must first eliminate all other possible causes or explanations for changes in the dependent
Confounding Variable
is a variable related to an independent variable that provides an alternative explanation for the relationship between the independent and dependent variables
Non Experimental Research Methods
non-experimental research methods are research methods designed to measure naturally occurring relationships between variables without having the ability to infer cause-effect relationships
some of the most common types of nonexperimental research designs include quasi-experiments, survey research, observational research, and archival research
Quasi-experimental research
compares naturally formed or preexisting groups rather than employing random assignment to conditions
