Concepts
Measurement Scales
- Nominal scales: These are used to labeling variables that are
of no quantitative value. For eg: shirt color: Red, Blue,
Black
- Ordinal scales: In the case of ordinal scales the order is
important but we don't know the differences between each of them.
For eg: 1-Not satisfied, 2- No effect, 3- Satisfied. The order is
important but we can't say how much satisfied is 3 compared to
1
- Interval: In this case, we know the order and also the
difference but we don't have a true zero. For eg, temperature
- Ratio: We have an order, difference and a true zero as
well
Sampling Methods
- 1. Random
sampling is a procedure of drawing random elements from the
population. Each element of the population has an equal probability
of getting selected. Should be used but difficult to implement in
practical
- 2. Systematic
sampling: In this sampling, we count and number every element of
the population and then take an element after a fixed interval. For
eg, Randomly choose a starting point and then selecting every pth
element from the population. It is easier than random
sampling
- 3. Convenience
sampling: This is the easiest sampling technique. We take whatever
the data is available to us. This should not be used unless we
don't have any other option left or we are in need of urgent
analysis
- 4. Cluster
sampling: In this sampling, we divide the population into
geographical (often) groups which are called clusters. After
grouping, we select any group randomly and take all the elements of
that group. This is also called as Block sampling
sometimes
- 5. Stratified
sampling: Here also, we divide the population into groups, called
strata but the grouping is done based on some common
traits. After strata division, we take samples from all
the strata by any of the sampling methods described
above
Type of study
- In an observational study, we collect data
based infer some conclusions based on the collected data. We don't
interfere with the subjects or affect the variables in any way.
There is no convincing evidence to claim a cause and effect
relationship
- While in an experiment, we group subjects and
apply some treatment to one group and leave another group without
treatment and then analyze the results. One of the major goals of
conducting an experiment is to check for a causal effect between
the two variables
- In correlational studies, we look for associations among
naturally occurring variables without interfering with the
variables.
Solution
This study was conducted to explore whether perception of money
by preschool age children is influenced by their family wealth.
This was investigated by testing how the children would recall the
size of a coin, with the assumption that the size recalled related
to the value placed on the coin.
To conduct the study, three local day-care facilities specializing
in preschool age children were randomly selected. Consent forms
were sent home to the parents of the kids, which also asked parents
for their average yearly household income. Since everyone from the
three selected facilities was used in the study, this is
cluster
sampling. The income data collected was
quantitative
data at the ratio level of
measurement.
From the parents who consented, the children were divided into two
groups: those whose family income was greater than $40,000, and
those whose family income was less than $40,000.
Each child was asked to draw a nickel. The resulting circle?s
diameter was then measured. When the shape drawn was not a perfect
circle, the largest and smallest diameters were averaged. The coin
diameter data was qualitative data at the
ratio level of
measurement.
The gender of each child was also collected. This data was
qualitative
data at the nominal level of
measurement.
Because this data was being collected at one point in time, this is
a terminated
study. Since we are not imposing a treatment on the children, this
is not an experiment
It is hypothesized that children from lower income families would
draw larger coins than children from higher income families.
"Please let me know if you have any doubts. Happy to help.
Please thumbs up if you like the solution. Thanks"