Practicum 7:

Describing data and Sample

information

Lecture 1 - Stangor Chapter 6

Lecture 2 - Stangor Addendum B
 Paradigms
Research Process

CONTEXT

Research Conceptua- Operationa- Research

question lisation lisation Design
 Two main methodologies

Qualitative approach Quantitative approach

Formal measures (data =
Capturing richness and depth
numbers)
(data =words)
Statistical analyses to make
Understand, describe, explore
comparisons or look for
phenomena
relationships
 Quantitative Research Designs
Type Type Type
Experimental Pre-Experimental Non-experimental

Design
Design Design
Random Two-Group
One shot case At One Point
study Correlational
Design Criterion
Cross-sectional
Random Multi-
Group Design
Design Pre-Post Test
Design Longitudinal
Random Pre-Post Panel
Test Cohort
Trend
Design
Matched Groups
 Sampling methods
Strategy employed to obtain research participants

Probability sampling Non-probability sampling

✓ Chance of being selected ✓ Probability of being selected
can be determined cannot be determined
What we achieved

Conceptual Measured
NUMBERS
variable variable
Scales of
Measurement
(Revision)
Nomina
Ordinal
l

Interval Ratio
 Scales of Measurement

Nominal Ordinal
Categories Rank
I.e. gender, religion 1st, 2nd, 3rd
Mutually exclusive More than previous

No specified distance
Variables are discreet: between them
nothing in-between male Variables are discreet -
and female nothing between 1st & 2nd
place
 Scales of Measurement

Interval Ratio
Underlying continuum
Underlying continuum
E.g. temperature
E.g. length
15.2° or 15.5° or 15.9°.
Rarity in social sciences

BUT there is no absolute zero

Has an absolute zero
If it is 0° it does not mean that
there is no temperature, it If you ran 0 km, it means
means it is cold you did not run at all – it is
the absence of distance
Descriptive statistics
Fundamentals of descriptive
statistics
• Data/numbers is used to answer your research question
• This is done through use of statistical analysis
• Statistics make sense of data and help us to interpret the data
• Population data is referred to as parameters
• Sample data is referred to as statistics
• Samples statistics is used to estimate population parameters
• Two branches of statistics: Descriptive & Inferential statistics
• Descriptive statistics either numerical or graphical
• Today: Frequency distributions and graphs
 Fundamentals
Statistical notation
⦿ X (X1) and Y (Y2) refers to measured variables

⦿ N refers to sample size / number of participants

⦿ Σ indicates that a set of scores should be summed (added up)

⦿ Rounding = when the whole computation is complete, the

answer is often rounded to two decimal places.

Example 1: Round 0.0234 and 0.0254 = 0.02 and 0.03

Example 2: Round 0.998 = 1
Did these students do well?
Test marks of 30 students
Maximum Score 50

46 21 32 25 43 12 23 18
13 21 7 40 17 16 20 17
20 15 19 27 10 24 9 23
15 12 8 23 27 19
Grouped
frequency
distributions (GFD)
Summarizing a set of numbers in a table to
have a visual picture
 Grouped frequency distributions
Class interval Tally Frequency

45 – 49 | 1
40 – 44 || 2
35 – 39 0
30 – 34 | 1
25 – 29 ||| 3
20 – 24 |||| ||| 8
15 – 19 |||| ||| 8
10 – 14 |||| 4
5– 9 ||| 3
N = 30
 Grouped Frequency Distribution:
Example
Test marks of 30 students:
46 21 32 25 43 12 23 18
13 21 7 40 17 16 20 17
20 15 19 27 10 24 9 23
15 12 8 23 27 19

Compile a frequency distribution with a class interval size of 5.

STEP 1: Determine the lowest and the highest score

The intervals will be multiples of 5.
Must include the lowest score and the highest score
THEREFORE in this case
Must be 0, 5, 10, 15 etc
And must include 7
 Grouped Frequency Distribution:
Example (Continue …)

STEP 2: Draw up three columns:

Class interval Tally Frequency

STEP 3: Fill in your class intervals, until the highest

score is included
STEP 4: Fill in the tallies
STEP 5: Indicate N.
 What about in-between values?

Such as 9.7 and 10.1?

 Grouped Frequency Distribution:
Important concepts
APPARENT TRUE
• Apparent upper limit = the • True UPPER limit = Σ (apparent
highest score included in the upper limit and the next
class interval interval’s apparent lower
limit)÷2
• Apparent lower limit = the = (14 +15)/2
lowest score included in the = 14.5
class interval
• True LOWER limit = Σ
(apparent lower limit and the
next interval’s apparent upper
limit)÷2
= (9 +10)/2
= 9.5
 Grouped Frequency Distribution:
Important concepts
Range = True upper limit of highest score – True lower limit of the lowest
score.
Thus 46.5 – 6.5 = 40 [OF (46 – 7) + 1 = 40]

Class interval size = True upper limit – True lower limit

= 14.5 - 9.5
= 5

Class midpoint = (class interval size ÷ 2) + True lower limit

= (5/2) + 9.5
= 12
Mode = midpoint of interval with highest frequency
 Example of a class interval

(110, 111, 112, 113, 114)

✓ Apparent upper limit
✓ Apparent lower limit
✓ True/Real upper limit
✓ True/Real lower limit
✓ Size of class interval
✓ Midpoint
Graphs
Three types of graphs
Bar Chart Histogram

Polygon
Match your data to your graph
Discreet / Continuous Data
Nominal – categories between values

Use Bar Graph Use Histogram

• Favorite car • Number of cars that pass a
• Gender
robot/hour
• Activities
• Number of children in a family
Use Frequency Polygon
• Weight
• Fractions
• Can be
negative
or positive
Does this
look familiar?

Type of sport Tally Frequency

Soccer IIII IIII IIII 14
Tennis IIII III 8
Running IIII IIII IIII III 18
Swimming IIII IIII 10
N= 50
Bar chart (nominal data)
 Type of Tally Frequency
sport
Soccer IIII IIII IIII 14

Bar chart Tennis

Running
IIII III
IIII IIII IIII III
8
18
Swimming IIII IIII 10
1. Draw a x-axis and y-axis N= 50
2. X-axis has your categories
3. Y-axis has your frequency
4. Divide the axis into equal units (include 0) – use your highest frequency as a
guideline
5. Name each axis
6. Indicate the frequencies for each category – they should not touch
7. Draw a bar for each category and label (soccer, tennis, etc.)
8. Titles for graphs at the BOTTOM of the graph
a) What type, of who, how many?
 Histogram
(discreet data = only whole numbers)
 Class interval Frequency

20 - 24 6

Histogram 15 - 19 18
10 - 14 12
5 - 9 16
N= 52

1. Construct the x-axis (class midpoints) and y-axis (frequency), include 0

2. Name each axis
3. Calculate the midpoints
4. Indicate the frequency of each midpoint
5. Draw a bar for each midpoint, the bars are adjacent (i.e. next to each other) to
indicate the variable is quantitative
6. Title at bottom of graph
a) What type, of who, how many?
Frequency curve/polygon
Frequency table of class attendance of 52
Psychology students
CLASS INTERVAL FREQUENCY (F) CLASS MIDPOINTS
25 - 29 0 27

20 - 24 6 22

15 - 19 18 17

10 - 14 12 12

5–9 16 7

0–4 0 2

N=52
Relationship between raw data,
frequency table and graph

Create a
Use raw data frequency Construct a
to … table that is graph
used to…

Source: Doing Social Research - Chapter 14: Mentz & Botha (2012)
THE SHAPE OF A
DISTRIBUTION
Skewed or normal
SKEWED DISTRIBUTIONS
SKEWED DISTRIBUTIONS