Lecture Notes

in
MLS 3105
(introduction to Medical Laboratory Science Research)

Module 5

Prepared by:

Joy G. Raso, PhD .

First Semester 2020

“The LORD is good, a strong refuge when troubles comes. He is

close to those who trust in Him”
Nahum 1:7
 DATA ANALYSIS AND INTERPRETATION, WRITING THE
RESEARCH PROPOSAL

Topics:
5.1 Data analysis
5.1.1 differential analysis
5.1.2 inferential analysis
5.1.2.1 univariate
5.1.2.2 bivariate
5.2.2.3 multivariate
5.2 Scales of measurement
5.2.1 ratio
5.2.2 interval
5.2.3 ordinal
5. 2.4 nominal
5.3 Measures of central tendency
5.3.1 mean
5.3.2 median
5.3.3 mode
5.4 Steps in data interpretation
5.5 Qualitative Data analysis
5.5 Research proposal definition
5.5.1 characteristics of a well written research
proposal
5.5.2 Parts of a research Proposal

• process of summarizing trends and patterns observed in the data, determining

major differences and relationships among variables used in the study and the
application of appropriate statistical tests on a set of data to answer the
objectives of a study.
-David,2002
• involved the computation of differentials and indexes used in determining
relationships between and among variables used in the study.

-Parel et al.,1979
Level/Scales of measurement:
1. Nominal
qualitative only.
Characterize by data that consist of names, labels, or qualities & categories
No mathematical computations can be made at this level
The data cannot be arranged in order just as low to high
no arithmetic origin, order, direction, or distance relationship
Would help answer the question: Is A different from B?
Example:
Sex: 1-Male 2- Female
Color of the Philippine flag: 1- Red, 2-yellow, 3-blue, 4-white
2. Ordinal
are either qualitative or quantitative.
Arranged in order, but differences between data entries are not meaningful.
Can be ranked they do not allow for relative degree of difference between them
or comparisons are not possible.
simply shows order of magnitude
Would help to answer the question: How is A bigger than B?
Example:
Top 3 students A= 94 B= 89 C=88

3. Interval
are quantitative.
A zero entry simply represents a position on a scale; the entry is not an inherent
zero.
Arranged in order, the differences between data entries can be calculated
Allows for degree of difference between items but not the ratio between them.
Would help to answer the question: By how many units do A and B differ?
Example:
Temperatures in Celsius scale 400C & 500C

4. Ratio
are similar to the interval level, but a zero entry is meaningful.
Permits the comparison of differences of values
aka scale with a fixed zero value
All mathematical operations are possible
Would answer the question: How many times A is bigger than B?
Example:
Monthly income A=20,000, B=30,000
Types of data analysis
1. DESCRIPTIVE STATISTICS
describe the nature and characteristics of an event or a population under
investigation
provide simple summaries and graph the data about the sample and the
measures
help us to simplify large amounts of data in a sensible way.
Example:
Suppose we want to describe the test scores in a specific class of 30 students. We
record all of the test scores and calculate the summary statistics and produce graphs

These results indicate that the mean score of this class is 79.18. The scores
range from 66.21 to 96.53, and the distribution is symmetrically centered
around the mean. A score of at least 70 on the test is acceptable. The data
show that 86.7% of the students have acceptable scores.
2. INFERENTIAL ANALYSIS
concerned with making conclusions
help us decide, whether the differences between groups that we see in our data
are strong enough to provide support for our hypothesis that group differences
exist in general, in the entire population.
analysis used in hypothesis testing to draw conclusions from a sample and
generalize them to a population
Example:
You randomly select a sample of 12th graders in the City and collect data on
their SAT scores and other characteristics.
You can use inferential statistics to make estimates and test hypotheses about
the whole population of 12th graders in the Region based on your sample data.

With descriptive statistics you are simply describing what is or what the data
shows.
In descriptive statistics, we picked the specific class that we wanted to describe
and record all the test scores for that class.
For inferential statistics, we need to define the population and then draw a
random sample from that population.
With inferential statistics, you are trying to reach conclusions that extend
beyond the immediate data alone. Inferential statistics is used to make
inferences from our data to more general conditions.
A. Describing the Characteristics of the Data UNIVARIATE ANALYSIS
involves the examination across cases of one variable at a time.

When the researcher wants to describe the characteristics of a same

population the following are use:

1. FREQUENCY DISTRIBUTION
indicates the number and percentage of responses for each category.
useful measure for analyzing nominal and ordinal data.
Data may be presented in table or graphical form
Example.
Table 1. Distribution of students according to sex

2. MEASURES OF CENTRAL TENDENCY

enable the researcher to summarize the data in a single number
it commonly called averages
a single value that is used to identify the “center” of the data
thought of as a typical value of the distribution
precise yet simple
most representative value of the data it is an estimate of the “center” of a
distribution of values.

1. MEAN
most popular, arithmetic average of all measurements in the population.
aka “average”
easy to compute
may not be an actual observation in the data set
can be applied in at least interval level
it is equal to the sum of all the values in the data set divided by the number
of values in the data set
most used method of describing central tendency. To compute the mean
all you do is add up all the values and divide by the number of values. For
example, the mean or average quiz score is determined by summing all
the scores and dividing by the number of students taking the exam.

Example, consider the test score values:

15,20,21,20,36,15,25,15 Mean= 167/8 = 20.875 or 20.88

2. MEDIAN
divides the observation into two equal parts
the middle/midpoint score for a set of data that has been arranged in order
of magnitude
less affected by outliers and skewed data
 can be applied in at least ordinal level

Example:
15,15,15,20,20,21,25,36 Median = 20

3. MODE
occurs most frequently occurring value in the set of scores
the most common data point
can be used for quantitative as well as qualitative
not affected by extreme values

Example:
82 95 85 93 87 96 90 91 93 87 94 83 95 95 88 Mode = 95

When the researcher wants to describe the variances in the data the
commonly use measures of variations are:

1. RANGE
simple measure of variation calculated as the highest value in the distribution
minus the lowest value plus 1
Range = highest value – lowest value + 1
97- 82 = 15 +1 =16

2. STANDARD DEVIATION
gives the average of the distances of individual observation from the group
mean, the square root of the average squared deviation of each case from
the mean

If there is a large amount of variation, then on average, the data values will
be far from the mean. Hence, the SD will be large.
If there is only a small of variation then on average, the data values will be
close to the mean. Hence, the SD will be small.

B. Analyzing Differences within the data

1. Differences in proportions
Make used of Z-test for difference in proportion.

Z-test
Is a statistical test to determine whether two population means are
different when the variances are known, and the sample size is large.
It can used to test hypotheses in which the z-test follows a normal
distribution.
 Z 1.96 (95%) 2.58 (99%)
p 0.05 (95%) 0.01 (99%)

Z-Score Table
is a table that shows the percentage of values (or area percentage) to the left of
a given z-score on a standard normal distribution.
Use the negative Z score to find values on the left of the mean as can be
seen in the graph alongside. Corresponding values which are less than the
mean are marked with a negative score in the z-table and respresent the
area under the bell curve to the left of z.
Use the positive Z score to find values on the right of the mean as can be
seen in the graph alongside. Corresponding values which are greater than the
mean are marked with a positive score in the z-table and respresent the area
under the bell curve to the right of z probability area.

Example:
Find the probability that a single person (sample size n = 1) gets a test
value BETWEEN 74 and 78 for their exam score. Assume that the exams are
normally distributed with a mean of 82 and a standard deviation of 4.

z = (x – mean)/SD
z = (74 – 82) / 4 = – 8/4 = -2
Therefore, the value 74 in our problem is the same as z = -2
z = (x – mean)/SD
z = (78 – 82)/ 4 = -4/4 = -1
So, the z value for 78 is -1
using the z-table determine the z-value in the probability area
p ( z<-2)= 0.0228
p(x<-1) = 0.1587

Normal Distribution/Gaussian Curve

The mean, median, and mode are all located at the center of the distribution
w/c is also the peak of the distribution
No skewed/normal
symmetric, unimodal, and asymptotic, perfectly symmetrical
 Location of the measures of central tendency for a distribution which is
skewed to the LEFT. Negatively skewed

Location of the measures of central tendency for a distribution which is

skewed to the RIGHT. Positively skewed

2. Z-test differences between Means

used to determine whether two samples' means are different when variances
are known, and sample is large (n ≥ 30)

Example:
A study comparing the performance of male and female college students
revealed that the 129 sample male students obtained a mean grade of 82.34
( SD =1.69), while 178 sample female students obtained a mean grade of
81.85 (SD =1.34). To determine whether the mean grade of the male students
significantly differ from that of the female students, the z-test difference
between means can be applied.

There is no significant difference between two means

(z= 1.74 p= .350)
C. Describing Relationships between variables (Bivariate Analysis)
bivariate analysis, where exactly two measurements are made on each
observation.
The two measurements will be called X and Y. Since X and Y are obtained
for each observation, the data for one observation is the pair (X,Y).
 It is the analysis of the relationship between the two variables.
- Bivariate data can be stored in a table with two columns

Example:

Four Aspects of a relationship association between variables

(Schutt,1999)
1. EXISTENCE
association between two variables exists when the percentages among
categories of the DV vary
2. STRENGTH
association between two variables is strong if there are big variations in the
percentage between or among categories of the DV
3. DIRECTION
for quantitative variables (interval & ratio) when the values of IV tend to
increase with increase in DV (positive) but when the values of IV tend to
increase with decrease DV (negative)
4. PATTERN
changes in the percentage of distribution of the DV may be regular or
not regular

To illustrate the fours aspects of relationship let us use the example below.
Table 1 Distribution of respondents by voting behavior and Age

Table 1 shows association exist between age and voting behavior of

below 25 and 25 and above. The data shows that there is a higher percentage
of voters among age below 25 (70%) than among those 25 and above (30%).
It is also obvious that the association between age and voting behavior is
strong because of the big difference between the percentage. The direction
of the association between variables is negative which means that younger
respondents are more likely to vote than older respondents.
Moreover, table 1 reveals a monotonic pattern because as the value of
one variable decreases the value of the other increases.
D. Multivariate Analysis
analysis of one variable on another when one or more are held constant
Allows determination of the effect variables on the relationship between two
variables controlling for one or more variables.

Example:
Factor Analysis = is a way to condense the data in many variables
into a just a few variables. For this reason, it is also sometimes called
“dimension reduction.”
Multivariate analysis of variance (MANOVA) = has two or
more independent variables and two or more dependent
variables.

DATA INTERPRETATION
process of explaining the meaning of the data in table with emphasis on the
highlights and trends shown by the data.
In this chapter, the captured data from the qualitative and quantitative
research is presented, analyzed, described, and interpreted in a systematic
manner as the next step of the research process

Steps in Interpreting Data

1. Review objectives, hypothesis, and theoretical/conceptual framework
2. Describe the data (percentage, fractions, ratios)
3. In describing and interpreting data focus on extreme numbers
(highest or lowest)
4. Data serve as supporting facts to the text, thus the table must be
placed after the text that presents and interprets the data in the table.
5. In interpreting cross-tabulated data, the researcher should describe or
discuss the DV in relation to IV.
6. If a hypothesis is being tested, it should be stated that the hypothesis is
either rejected or not rejected
7. Studies with theoretical/conceptual framework must show whether or
not the data support the theory used.
8. The findings of related studies should be compared with the results of
the study. Do the findings contradict or support findings of the previous
studies? What are the contraindications? What are the possible reasons for
contradictions?
QUALITATIVE DATA ANALYSIS
Results of key informant interview and FGDs are summarized,
Categorized and described
Qualitative data are useful supplement for quantitative data and they can
also provide detail to the statistical data and analysis.
In addition to descriptions, relevant direct quotes of the rrespondents may
also be included in the analysis.
Unique situations/views or voices of the respondents or key informants can
be directly quoted and included in the ddiscussion to give “color and life” to
the data.

WRITING THE RESEARCH PROPOSAL

A research proposal is a plan of how study will be undertaken
It is describing the steps that will be followed in the conduct of the
study from problem identification to analysis.

Characteristics of a well-written research proposal:

1. Persuasive
2. Complete
3. Clear
4. Flexible

1. Title
2. Introduction
3. Statement of the problem
4. Theoretical and Conceptual Framework (if needed)
5. Hypothesis (if any)
6. Significance of the study
7. Scope and limitation
8. Review of related literature
9. Methodology
10. References
11. Schedule of activities (Gantt Chart)
12. Budget
13. Appendices (research instruments)
Chapter I- INTRODUCTION
- Background and Rationale of the study
- Objectives of the study
- Hypothesis (if any)
- Theoretical and Conceptual Framework (if needed)
- Conceptual and Operational definitions of variables
- Significance of the study
- Scope and limitation of the study
Chapter II- REVIEW OF RELATED LITERATURE
Chapter III- METHODOLOGY
- Research Design
- Population element/Respondents and Sampling procedure
- Data Collection
- Data processing
Chapter IV DATA ANALYSIS AND INTERPRETATION
- Summary of findings
- Recommendations
REFERENCES
APPENDICES
- Research instruments
- Dummy tables
- Schedule of activities/Gantt chart

Writing the Final Research Report

Major parts
Preliminary sections
- title page
- abstract
- approval sheet
- table of contents
- list of tables/figures
- acknowledgement
Body
- Chapter I
- Chapter II
- Chapter III
- Chapter IV
References
Appendices
GANTT CHART
is a type of bar chart that illustrates a project schedule, named after its
inventor, Henry Gantt, who designed such a chart around the years 1910–
1915. Modern Gantt charts also show the dependency relationships between
activities and current schedule status

Citations are important because they show your reader(s):

- when the information was published
- who the author of the information is
- which journal or group published the information
- which version the information was published in (usually just for literature)

Most common citation styles:

- APA (The American Psychological Association)
- MLA (Modern Language Association).
- Chicago (published by the University of Chicago)
- CSE (Council of Science Editors)
- IEEE (Institute of Electrical and Electronics Engineers).