Graphs and Charts Flashcards
What are graphs and charts used for?
Summarizing results visually
What do scatter graphs show us?
the relationship between two variables plotted on x and y axis
What can we identify from the relationship of the two variables shown on a scatter graph?
correlational trends, e.g. height and weight are positively correlated
What is a coefficient used for?
To measure the strength and direction of correlations
What do coefficients range from?
+1.0 to -1.0
what would a coefficient of ….. mean?
-0.3
-0.8
+0.3
+0.8
Weak Negative
Strong Negative
Weak Positive
Strong Positive
What can correlational interpretations not predict?
Cause and effect
What are correlational coefficients used for?
Bi-variant data
Correlational analysis
What type of data is a bar graph used for?
Univariant data
What does a bar graph show?
The frequency of data
A bar graph is the best way of displaying….what?
nominal data that has been collected into categories
What does the highest bar of bar graph show?
Modal variable
A gap in the bar means the data is… ?
categorical/nominal data
- Means no link across x axis
- Known as discreet data
Strength of a bar graph
You can compare multiple categories on one chart
What is a histogram?
Very similar to a bar graph but for frequency of data = CONTINUOUS DATA
No gaps in bars shows continuous relationship scale
Strengths of a frequency polygon, made from a histogram (2)
better for showing trend of continuous data
Still be used to compare multiple sets of data
Interpreting significance: what does p≤0.005 mean?
The probability that the results actually occurred by chance are 5% or less so the results are significant
What does p>0.005 mean?
Means the probability of the results happening by chance are grater than 5% so the results are not significant
What is a Type 1 error?
A Type I Error is also known as a False Positive or Alpha Error. This happens when you reject the Null Hypothesis even if it is true. The Null Hypothesis is simply a statement that is the opposite of your hypothesis. For example, you think that boys are better in arithmetic than girls. Your null hypothesis would be: “Boys are not better than girls in arithmetic.”
You will make a Type I Error if you conclude that boys are better than girls in arithmetic when in reality, there is no difference in how boys and girls perform. In this case, you should accept the null hypothesis since there is no real difference between the two groups when it comes to arithmetic ability. If you reject the null hypothesis and say that one group is better, then you are making a Type I Error.
What is a Type 2 error?
A Type II Error is also known as a False Negative Error. This happens when you accept the Null Hypothesis when you should in fact reject it. The Null Hypothesis is simply a statement that is the opposite of your hypothesis. For example, you think that dog owners are friendlier than cat owners. Your null hypothesis would be: “Dog owners are as friendly as cat owners.”
You will make a Type II Error if dog owners are actually friendlier than cat owners, and yet you conclude that both kinds of pet owners have the same level of friendliness. In this case, you should reject the null hypothesis since there is a real difference in friendliness between the two groups. If you accept the null hypothesis and say that both types of pet owners are equally friendly, then you are making a Type II Error.
