Allama Iqbal Open University Islamabad

Book name
(8614)
Level: B.Ed.

Name: Saba Usman

Roll No. BY627929
Semester: Autumn 2020
Assignment No.1
Question No.1
Describe level of measurement. Give five examples of each level and explain the role of level
of measurement in decision-making.
Answer:
Level of Measurement
There are two basic types of variables - quantitative and categorical. Each uses different type
of analysis and measurement, requiring the use of different type of measurement scale. A scale
of a variable gives certain structure to the variable and also defines the meaning of the variable.
There are four types of measurement scales: nominal, ordinal, interval, and ratio.
Nominal Scale
A nominal scale is the simplest form of measurement researchers can use. The word nominal
means "having to do with names. “Measurements made on this scale involve merely naming
things. It consists of a set of categories that have different names. Such measurements label
and categorize observations but do not make quantitative distinctions between them. For
example, if we wish to know the sex of a person responding to the questionnaire, we would
measure it on nominal scale consisting of two categories (male or female). A researcher
observing the behavior of a group of infant monkeys might categorize responses as playing,
grooming, feeding, acting aggressively or showing submissiveness. As the researcher merely
gives names to each category so, this is a nominal scale of measurement. The nominal scale
consists of qualitative distinctions.
Although, a nominal scale consists of qualitative differences, yet it does not provide any
information about quantitative differences between individuals. Numerical values like 0 and 1
are merely used as code for nominal categories when entering data into computer programs.
For example, a fund manager may choose to assign the number 1 to small-cap stocks, the
number 2 to corporate bonds, the number 3 to derivatives, and so on.
Ordinal Scale
In ordinal scale of measurement, the categories that make up the scale not only have separate
names but also are ranked in terms of magnitude. This scale consists of a set of categories that
are organized in an ordered sequence. For example, a manager of a company is asked to rank
employees in term of how well they perform their duties. The collected data will tell us who
the manager considers the best worker, the second best, and so on. The data may reveal that
the worker, who is ranked second, is viewed as doing better work than the worker who is ranked
third. However, we can get no information about the amount that the workers differ in job
performance, i.e. we cannot get the answer of the question "How much better?" Thus, an
ordinal scale provides us information about the direction of difference between two
measurements, but it does not reveal the magnitude of the difference.
For example, a list of 500 managers of mutual funds may be ranked by assigning the number I
to the best-performing manager, the number 2 to the second best-performing manager, and so
on. With this type of measurement, one can conclude that the number I-ranked mutual fund
manager performed better than the number 2-ranked mutual fund manager.
Interval Scale
An interval scale possesses all the characteristics of an ordinal scale, with additional feature
that the categories form a series of intervals that are exactly of the same size. This additional
information makes it possible to compute distances between values on an interval scale. For
example, on a ruler I-inch interval is the same size at every location on the ruler. Similarly 4-
inch distance is exactly the same size no matter where it is measured on the ruler. Similarly,
the distance between the scores of 70 and 80 is considered to be the same as the distance
between scores of 80 and 90. For all practical purposes these numbers can undergo arithmetic
operations to be transformed into meaningful results. Interval scale answers the question "How
much better?" or "How much is the difference?" But there is no intrinsic zero, or starting point.
The zero point on the interval scale does not indicate a total absence of what is being measured.
For example, 00 (zero degree) on the Celsius or Fahrenheit scale does not indicate no
temperature. For example, temperature measurement is an example of an interval scale: 60°C
is colder than 65°C and the difference in temperature is the same as the difference between
50°C and 55°C. In other words, the difference of 5°C in both intervals have the same
interpretation and meaning. Consider why the ordinal scale example is not an interval scale: A
fund manager that is ranked I probably did not outperform the fund manager ranked 2 by the
exact same amount that a fund manager ranked 6 outperformed a fund manager ranked 7.
Ordinal scales provide a relative ranking, but there is no assurance that the differences between
the scale values are the same. A drawback in interval scales is that they do not have a true zero
point. Zero does not represent an absence of something in an interval scale. Consider that the
temperature -O°C does not represent the absence of temperature. For this reason, interval-seale-
based ratios fail to provide some insights - for example, 50°C is not twice as hot as 25°C,
Ratio Scale
A ratio scale has all the characteristics of an interval scale but adds an absolute zero point. It
means on a ratio scale a value of zero indicates complete absence of the variable being
measured. Advantage of absolute zero is that a ratio of numbers on scale reflects ratio of
magnitude for the variable being measured. We can say that one measurement is three times
larger than another, or one score is only half as large as another. Thus, ratio scale not only
enables us to measure the difference between two individuals, but also to describe the
difference in terms of ratios.
For example, the measurement of money is an example of a ratio scale. An individual with $0
has an absence of money. With a true zero point, it would be correct to say that someone with
$100 has twice as much money as someone with $50.
Question No.2
Differentiate between primary and secondary data. Give meaningful examples with
explanation.
Answer
1. Meaning, example, and definition
Primary data are fresh (new) information collected for the first time by a researcher himself for
a particular purpose. It is a unique, first-hand and qualitative information not published before.
It is collected systematically from its place or source of origin by the researcher himself or his
appointed agents. It is obtained initially as a result of research efforts taken by a researcher
(and his team) with some objective in mind. It helps to solve certain problems concerned with
any domain of choice or sphere of interest. Once it is used up for any required purpose, its
original character is lost, and it turns into secondary data.
Secondary data, on the other hand, are information already collected by others or somebody
else and later used by a researcher (or investigator) to answer their questions in hand. Hence,
it is also called second-hand data. It is a ready-made, quantitative information obtained mostly
from different published sources like companies' reports, statistics published by government,
etc. Here the required information is extracted from already known works of others (e.g.
Published by a subject scholar or an organization, government agency, etc.). It is readily
available to a researcher at his desk or place of work.
2. Data's originality
Primary data are collected by a researcher (or investigator) at the place or source of its origin.
These are original or unique information.
A researcher (or investigator) does the collection of secondary data from already existing works
of others. These are neither originals nor unique information.
3. Need of adjustment
The primary data collection is done to accomplish some fixed objective, and obtained with
some focus in mind. Hence, it doesn't need any prior adjustment before getting used to satisfy
the purpose of an inquiry.
Secondary data collected are truly the work of someone else done for some other purposes. It
is not focused to meet the objective of the researcher. As a result, it needs to be properly
adjusted and arranged before making its actual use. Only after proper adjustment, it can be
accustomed to some extend for achieving the aim of a researcher.
4. Type of data
Primary data provide qualitative data. It means it gives information on subjective quality-
related features like look, feel, taste, lightness, heaviness, etc., of any object or phenomenon
under research or inquiry.
On the contrary, secondary data, provide quantitative data. In other words, it gives information
about an object or event in a numerical, statistical and tabulated form like in percentages, lists,
tables, etc.
6.0btained data's reliability
Primary data are more reliable than secondary data. It is because primary data are collected by
doing original research and not through secondary sources that may subject to some errors or
discrepancies and may even contain out-dated information. Secondary data are less reliable
than primary data. It is so, since, based on research work done by others and not by the
researcher himself. Here, verification of published information cannot be always confirmed
accurately as all references.
used may not be available or mentioned in detail.
7. Need of investigator
Collection of primary data needs availability of trained researchers or investigators. Further,
they also need to be adequately supervised and controlled. If the availability of trained
investigators and cost involved in hiring them is a problem, then in such a case, secondary
methods of data collection are recommended. Its data collection doesn't need to hire them.
8. Cost effectiveness
Primary data collection needs the appointment of a team that mainly comprises of researchers,
field interviewers, data analysts, so on. Hiring of these experts and other additional costs,
demands more funds to be allocated to complete research work on time. For this reason, it is a
costly affair. The secondary data collection doesn't require the appointment of such a team.
Here, since no experts hired, cost is minimized. As a result, it is very economical.

Question No.3
Explain advantages and disadvantages of bar charts and scatter plot.
Answer:
Advantages and Disadvantages of Bar Charts
Following are the advantages of bar charts.
i) They show data category in a frequency distribution.
ii) They display relative numbers / proportions of multiple categories.
iii) They summarize a large amount of data in an easily interpretable manner.
iv) They make trends easier to highlight than tables do.
v) By bar charts estimates can be made quickly and accurately.
vi) They are easily accessible to everyone.
Following are the disadvantages of bar charts.
i) They often require additional explanation.
ii) Thy fail to expose key assumptions, causes, impacts and patterns
iii) They can be manipulated to give false impressions.
Scatter Plots
Advantages of Scatter plots:
1. Show a relationship and a trend in the data relationship.
2. Show all data points, including minimum and maximum and outliers.
3. Can highlight correlations.
4. Retains the exact data values and sample size.
5. Shows both positive and negative type of graphical correlation.
Disadvantages of Scatter Plots:
1. Flat best-fit line gives inconclusive results.
2. Interpretation can be subjective.
3. Correlation does not mean and not show causation.
4. Data on both axes have to be continuous data (see our post discrete vs
continuous data).
5. You cannot use Scatter diagrams to show the relation of more than two variables

Question No.4
Explain normal distribution. How does normality of data affect the analysis of data?

Answer
Normal Distribution
Normal distribution, also known as the Gaussian distribution, is a probability distribution that
is symmetric about the mean, showing that data near the mean are more frequent in occurrence
than data far from the mean. In graph form, normal distribution will appear as a bell curve.
The normal distribution is the most common type of distribution assumed in technical stock
market analysis and in other types of statistical analyses. The standard normal distribution has
two parameters: the mean and the standard deviation. For a normal distribution, 68% of the
observations are within +/- one standard deviation of the mean, 95% are within +/- two standard
deviations, and
99.7% are within +- three standard deviations.
Effects of Normality of Data
The normality assumption is one of the most misunderstood in all of statistics. In multiple
regression, the assumption requiring a normal distribution applies only to the disturbance term,
not to the independent variables as is often believed. Perhaps the confusion about this
assumption derives from difficulty understanding what this disturbance term refers to simply
put, it is the random error in the relationship between the independent variables and the
dependent variable in a regression model Each case in the sample actually has a different
random variable which encompasses all the "noise" that accounts for differences in the
observed and predicted values produced by a regression equation, and it is the distribution of
this disturbance term or noise for all cases in the sample that should be normally distributed.
There are few consequences associated with a violation of the normality assumption, as it does
not contribute to bias or inefficiency in regression models. It is only important for the
calculation of p values for significance testing, but this is only a consideration when the sample
size is very small. When the sample size is sufficiently large (>200), the normality assumption
is not needed at all as the Central Limit Theorem ensures that the distribution of disturbance
term will approximate normality. When dealing with very small samples, it is important to
check for a possible violation of the normality assumption. This can be accomplished through
an inspection of the residuals from the regression model (some programs will perform this
automatically while others require that you save the residuals as a new variable and examine
them using summary statistics and histograms). There are several statistics available to examine
the normality of variables. including skewness and kurtosis, as well as numerous graphical
depictions, such as the normal probability plot. Unfortunately the statistics to assess it are
unstable in small samples, so their results should be interpreted with caution. When the
distribution of the disturbance term is found to deviate from normality, the best solution is to
use a more conservative p value (.01 rather than .05) for conducting significance tests and
constructing confidence intervals. More precisely, the tests are a form of model selection, and
can be interpreted several ways, depending on one's interpretations of probability:
6. l.In descriptive statistics terms, one measures a goodness of fit of a normal model to the data
- if the fit is poor then the data are not well modeled in that respect by a normal distribution,
without making a judgment on any underlying variable.
7. 2.In frequentist statistics statistical hypothesis testing, data are tested against the null
hypothesis that it is normally distributed.
8. 3.In Bayesian statistics, one does not "test normality" per se, but rather computes the
likelihood that the data come from a normal distribution with given parameters /1,(J (for all
/1,(J), and compares that with the likelihood that the data come from other distributions under
consideration, most simply using a Bayes factor (giving the relative likelihood of seeing the
data given different models), or more finely taking a prior distribution on possible models and
parameters and computing a posterior distribution given the computed likelihoods.
A normality test is used to determine whether sample data has been drawn from a normally
distributed population (within some tolerance). A number of statistical tests, such as the
Student's t-test and the one-way and two-way ANOV A require a normally distributed sample
population.

Question No.5
How is mean different from median? Explain the role of level of measurement in measure of
central tendency.

Answer:
Mean
Mean is the most commonly used measure in educational research. It is appropriate for
describing ratio or interval data. It can also be used for both continuous and discrete numeric
data. It is the arithmetic average of the score. It is determined by adding up all the scores and
then by the sum by the total number of scores. Suppose we have scores, 40, 85, 94, 62, 76, 66,
90, 59, 68, and 84. In order to find the mean of these scores we simply add all the scores, which
comes to 724. Then divide this sum 10 (total number of scores). We will get 72.4, which is the
mean score. The formula for computing the mean is: (Mean score) X = DUn Where L represents
"Sum of', X represents any raw score value, n represents total number of scores. We can also
define mean as mean is the amount each individual would get if the total (LX) were divided
equally among all the individual members (n) in the distribution. In some other words we can
say that the mean is the balance point for the distribution. To interpret the as the "balance point
or the center value", we can use the analogy of a seesaw. Its mean lies right at the center where
the fulcrum keeps the board perfectly balanced. As the mean is based on every score or value
of the dataset so it is influenced by outliers and skewed distribution. Also it cannot be calculated
for categorical data as the values cannot be summed.
Median
Median is the middle value of rank order data. It divides the distribution in two halves (i.e. 50%
of scores or observations on either side of median value). It means that this value separates
higher half of the data set from the lower half. The goal of the median is to determine the
precise midpoint of the distribution. Median is appropriate for describing ordinal data.
Procedure for Determining Median
When the number of scores is odd, simply arrange the scores in order (from lower to higher or
from higher to lower). The median will be the middle score in the list. Consider the set of scores
2, 5, 7, 10, 12. The score "T' lies in the middle of the scores, so it is median. When there is an
even number of scores in the distribution, arrange the scores in order (from lower to higher or
from higher to lower). The median will be the average of the middle two score in the list.
Consider the set of scores 4, 6, 9, 14 16, 20. The average of the middle two scores 11.5 (i.e. 9+
14/2 = 23/2 = 11.5) is the median of the distribution. Median is less affected by outliers and
skewed data and is usually preferred measure of central tendency when the distribution is not
symmetrical. The median cannot be determined for categorical or nominal data.

Central Tendency
Central tendency is a descriptive summary of a dataset through a single value that reflects the
center of the data distribution. Along with the variability (dispersion) of a dataset, central
tendency is a branch of descriptive statistics. The central tendency is one of the most
quintessential concepts in statistics. Although it does not provide information regarding the
individual values in the dataset, it delivers a comprehensive summary of the whole dataset.

Role of Measures of Central Tendency

Generally, the central tendency of a dataset can be described using the following measures:

Mean (Average):
Represents the sum of all values in a dataset divided by the total number of thevalues.

Median:
The middle value in a dataset that is arranged in ascending order (from the smallest value to
the largest value). If a dataset contains an even number of values, the median of the dataset is
the mean of the two middle values.

Mode:
Defines the most frequently occurring value in a dataset. In some cases, a dataset may contain
multiple modes while some datasets may not have any mode at all. Even though the measures
above are the most commonly used to define central tendency, there are some other measures,
including, but not limited to, geometric mean, harmonic mean, midrange, and geometric
median. The selection of a central tendency measure depends on the properties of a dataset. For
instance, the mode is the only central tendency measure for categorical data, while a median
works best with ordinal data. Although the mean is regarded as the best measure of central
tendency for quantitative data, that is not always the case. For example, the mean may not work
well with quantitative datasets that contain extremely large or extremely small values. The
extreme values may distort the mean. Thus, you may consider other measures. The measures
of central tendency can be found using a formula or definition. Also, they can be identified
using a frequency distribution graph. Note that for datasets that follow a normal distribution,
the mean, median, and mode are located on the same spot on the graph.