DATA

ANALYSIS
IN RESEARCH Presented by Group 4
Merilo, Katrina Terante, Roxanne Mae
Pacan, Venus Piolo Urbano, Ruella Jenn
Penaflor, Flare Anne Velasco, Shannille Everylle
Roja, Cindy Joy Yanson, Chelsae A.
Tafalla, Kristel
DATA ANALYSIS is an important phase in the
research, which includes:
- Computation of certain measures
- Searching for relationship that exist
among variables.

This part provides answers to your research

questions and hypothesis
Data Analysis interprets and analyzes the
collected data according to the study objectives.

This includes:
★ Compilation
★ Editing
★ Coding
★ Classification
★ Presentation of Data
DEFINITION: Process of organizing and
synthesizing the data as to answer the research
questions and test hypothesis

PURPOSE:
★ To describe the data in meaningful terms.
★ To analyze the data so that patterns of
relationships can be detected.
QUANTITATIVE
DATA ANALYSIS
 STEPS OF QUANTITATIVE DATA ANALYSIS

1. Data Preparation
● Compilation
● Editing
● Coding
● Classification
● Tabulation
 STEPS OF QUANTITATIVE DATA ANALYSIS

2. Describing the Data

● Descriptive statistics (e.g., percentage,
mean, median, mode)
3. Drawing the Inferences of Data
● Inferential statistics (e.g., relationships,
etc.)
 STEPS OF QUANTITATIVE DATA ANALYSIS

4. Interpretation of Data
● Examination of analyzed study results
● Draw inferences and conclusions
 01 SCALES OF
MEASUREMENT IN
QUANTITATIVE
STUDIES
 LEVELS OF MEASUREMENT
Nominal Ordinal Interval Ratio
Measurement Measurement Measurement Measurement
Numbers represent Ranks attributes in a Ranks with equal Highest level, includes all
categories (e.g., specific order. Intervals intervals between units interval properties plus a
gender: 1 = female, 2 = between ranks are not (e.g., temperature, IQ meaningful zero (e.g.,
male). No quantitative equal; order matters. scores).No true zero weight, height). Allows for
meaning; numbers are point; meaningful the full range of
labels only. •Example:I n an activities distances between mathematical operations,
of daily living scale, 1 values. including ratios.
•Example: Cannot might mean completely
calculate an average dependent, 2 means needs •Example: Difference •Example: 200 lbs is twice
for gender. assistance, and 4 means between 120 and 100 is as heavy as 100 lbs; 0
completely independent. the same as between means the absence of the
140 and 120 attribute.
 Descriptive Statistics
Descriptive Statistics- summarize and describe data.
Parameter - Index calculated from population data.
Statistic - Index calculated from sample data.
Key Characteristics
● Shape of Distribution: How values are spread out
(symmetric or skewed).
● Central Tendency: Where values cluster (e.g., mean,
median, mode).
● Variability: Spread of values (e.g., range, standard
deviation).
 Frequency Distributions
Organizes values from lowest to highest. Shows how often each
value occurs. Can be visualized using frequency polygons (graphs with
values on the x-axis, frequencies on the y-axis).
★ Unimodal Distribution - has one peak
★ Bimodal Distribution - has two peaks
★ Multimodal Distribution - has two or more peaks
Normal Distribution - Bell-shaped, symmetric, unimodal (common for
human attributes like height).
 Central Tendency
Mode Median Mean
is the number that is the point in a equals the sum of all
occurs most frequently distribution that values divided by the
in a distribution. In the divides scores in half. number of participants,
following distribution: Consider the following what we usually call the
50 51 51 52 53 53 53 53 set of values: average. Example: 85 109
54 55 56 22334567896 120 135 158 177 181 195

The mean is 53 The median is 5 The mean is 145

 Variability
Describes how spread out data is, shown in frequency distributions.
Variability can differ even if distributions share the same mean (e.g., two
schools with similar means but different score spreads).
Two Common Indexes:
● Range- Difference between the highest and lowest score in a
distribution.
● Standard Deviation- calculated based on every value in a
distribution. Summarizes the average amount of deviation of
values from the mean.
Bivariate Descriptive Statistics
Describes relationships between two variables.
Cross Tabulations (Crosstabs tables)- A
two-dimensional frequency distribution showing the
relationship between two variables.
•Example: Crosstabs can reveal patterns, such as
comparing gender and smoking status.
Correlation
Correlation describes the relationship between two
variables.

Correlation Coefficient: A statistic (ranging from -1.00 to +1.00)

that indicates the strength and direction of the relationship.
- Positive correlation: Both variables increase together.
- Negative correlation: One variable increases as the other
decreases.
Pearson’s r: Most commonly used for continuous
variables.

Continuous variable: Pearson’s r applies to

continuous variables, which are variables that can
take on a range of values. This is implied when
explaining what Pearson’s r is used for.
Absolute Risk (AR)
The proportion of people who experience an undesirable outcome in each group.
•Example: In a smoking cessation study, AR of continued smoking is 0.50 (50%) in the
intervention group and 0.80 (80%) in the control group.

Absolute Risk Reduction (ARR):

The difference in absolute risk between the unexposed (control) and exposed (intervention)
groups.
•Example: ARR = 0.80 (control) - 0.50 (intervention) = 0.30. This means 30% fewer smokers
would continue smoking with the intervention.

Odds Ratio (OR):

The ratio of the odds of an adverse outcome in the exposed group to the odds in the
unexposed group.
•Example: In the smoking cessation study, OR = 0.25, meaning smokers without the
intervention are 4 times more likely to continue smoking than those with the intervention.
Parameter Estimation
Estimates a population parameter (e.g., mean or proportion)
based on sample data.

•Point Estimation: A single value estimate.

•Interval Estimation: Uses Confidence Intervals (CI) to provide a
range of values with a specified probability.
○ 95% CI: Commonly used, indicates a 5% chance of being
wrong.
○ 99% CI: More conservative, with only a 1% chance of error
but a wider range of possible values.
Hypothesis Testing
A method to determine if observed differences in sample data
reflect true population differences or are due to chance (sampling
error).
★ Null Hypothesis (H₀): No effect or difference (e.g., the
intervention has no impact).
★ Research Hypothesis (H₁): The intervention is effective.
★ Hypothesis Testing Process - Compute a test statistic (e.g.,
comparing post-intervention means).Compare the test
statistic against a theoretical distribution.
Errors in Hypothesis Testing
•Type I Error: Rejecting a true null hypothesis (false-positive).

Example: Concluding the intervention worked when it didn’t.

•Type II Error: Accepting a false null hypothesis (false-negative).

Example: Concluding the intervention didn’t work when it did.

Level of Significance (Alpha, α)- The probability of making a Type I error

(rejecting a true null hypothesis).
Common Levels:
α = 0.05: 5% chance of a Type I error (analogous to a 95% Confidence Interval).
α = 0.01: 1% chance of a Type I error (analogous to a 99% Confidence Interval).
Steps in Hypothesis Testing
1.Select a Statistical Test - Based on variables’ level of measurement.
2.Specify Level of Significance (α) - Commonly set at 0.05.
3.Compute Test Statistic: Calculate based on study data.
4.Determine Degrees of Freedom (df) - Number of observations that can
vary in the analysis.
5.Compare to Theoretical Value - Use the calculated statistic to determine if
the result is significant or nonsignificant.

p-value- If the p-value is 0.025, it indicates that the result could have occurred
by chance only 25 times out of 1,000 (or 2.5%).
•Comparison to Alpha (α = 0.05): Since 0.025 < 0.05, the result is
statistically significant.
 BIVARIATE
STATISTICAL TEST
t-Test
● Researchers frequently compare two groups of people on
an outcome. A t-Test is a parametric test for testing
differences in two group means.

● TYPES:
○ Independent t-Test (two-sample t-Test or unpaired
t-Test)
○ Paired t-Test (Dependent t-Test)
○ One-sample t-Test
APPLICATION
● Medical or Clinical Research
○ Example: A clinical trial comparing the effectiveness of
a new drug versus a placebo.
● Educational Research
○ Example: Comparing the performance of two teaching
methods (traditional vs. e-learning).
● Psychological or Social Studies
○ Example: Comparing stress levels between two groups,
such as working adults vs. students.
t-Test Formula
Analysis of Variance (ANOVA)
● Analysis of Variance (ANOVA) is a statistical formula used
to compare variances across the means of different
groups. A range of scenarios use it to determine if there is
any difference between the means of different groups.
 APPLICATION
● Medical Research
○ Example: Comparing the effectiveness of three or more
different drugs on reducing cholesterol levels.
● Agriculture Studies
○ Example: Testing the effect of different fertilizers (3 or
more types) on crop yield.
ANOVA Formula
Chi-squared Test
● A chi-square test is a statistical test used to compare
observed results with expected results. The purpose of
this test is to determine if a difference between observed
data and expected data is due to chance, or if it is due to a
relationship between the variables you are studying.

● Two main types of chi-squared Test

○ Chi-Squared Test for Independence
○ Chi-Squared Goodness of Fit Test
APPLICATION
● Marketing Research
○ Example: Testing whether customer satisfaction levels
(satisfied/neutral/dissatisfied) vary based on the type
of product (A, B, or C).
● Social Sciences
○ Example: Investigating the relationship between
gender (male/female) and voting preference
(candidate A/candidate B).
Chi-squared Formula
Correlation Coefficients
● A correlation coefficient is a number between -1 and 1 that
tells you the strength and direction of a relationship
between variables. In other words, it reflects how similar
the measurements of two or more variables are across a
dataset. When one variable changes, the other variables
change in the same direction.

● Common Type of Correlation Coefficient

○ Pearson’s correlation coefficient - which measures the
degree of linear association between two variables.
APPLICATION
● Economics
○ Example: Examining the relationship between income
and household expenditure.
● Market Research
○ Example: Studying the relationship between
advertising expenditure and sales revenue.
● Environmental Studies
○ Example: Analyzing the relationship between
temperature and ice cream sales.
Correlation Coefficients
Formula
02 INTERPRETING
QUANTITATIVE
FINDINGS
OBJECTIVITY OF
FINDINGS
Objectivity means that research
findings are free from bias and
personal influence. The results
should reflect the true nature of what
was studied without being affected
by the researchers' opinions or
expectations.
DATA COLLECTION AND RESEARCH QUESTIONS
EXAMPLE:
If the research question asks, 'Does
The data collected in a study should daily exercise lower blood pressure?'
directly address the research the data collected should measure
questions posed by the blood pressure before and after the
researchers. This alignment is participants exercise. Collecting
essential for gathering relevant and unrelated data, like participants'
meaningful information. favorite foods, won’t help answer the
question.
 GENERALIZABILITY
Generalizability refers to the extent to
which research findings can be applied
to a larger population beyond the study
sample. It’s about whether the results
can be relevant to other groups or
settings.
CREDIBILITY OF RESULTS
Credibility pertains to how believable
and trustworthy the research findings
are. It is established through rigorous
methodologies and sound research
practices.
DETERMINING
PRECISION
Precision refers to the accuracy and reliability of
the measurements obtained in a study. It
indicates how closely the results reflect the true
values.
ANSWERING THE HYPOTHESIS:
SIGNIFICANT OR INSIGNIFICANT

Significant result: p ≤0.05

Insignificant results: p >0.05
QUALITATIVE
DATA ANALYSIS
01 Understanding the
Analysis of Qualitative
Data
 Qualitative Data
➔ Qualitative analysis is a challenging,
labor-intensive activity, with few fixed rules
➔ Qualitative analysis usually begins with
efforts to understand and manage the mass
of narrative data by developing a coding
scheme.
 Developing a Coding Scheme
1. Descriptive coding - uses mainly nouns as codes and is often used by beginning
qualitative researchers; it does not, however, provide much insight into meaning.
2. Process coding - often involves using gerunds as codes to connote action and
observable activity in the data.
3. Concept coding - involves using a word or phrase to represent symbolically a
broad meaning beyond observable factors or behaviors; the codes are usually
nouns or gerunds.
4. In vivo coding - involves using participant-generated words and phrases; it is
used as initial coding in many grounded theory studies.
5. Holistic coding - involves using codes to grasp broad ideas in large “chunks” of
data rather than coding smaller segments.
 “The other day, we ran out of everything
and we had to go to a church and get food.”
1. Descriptive coding
○ Code: food pantry use
2. Process coding
○ Code: dealing with food shortages
3. Concept coding
○ Code: coping with the risk of hunger
4. In vivo coding
○ Code:“The
ran out ofother
everythingday,
and wewe
had toran
go to aout
churchof
and get food.
everything and we had to go to a
church and get food.”
“I buy on deals. I learned how to, you know, what to buy and what not to buy. Where to
shop, where to look for sales. I’ll go to all the stores. And I clip coupons from the paper
and stuff. But sometimes that’s not enough. The other day, we ran out of everything
and we had to go to a church and get food.”

5. Holistic coding
Code: food management strategies

“I buy on deals. I learned how to, you know, what to buy and
what not to buy. Where to shop, where to look for sales. I’ll go to
all the stores. And I clip coupons from the paper and stuff. But
sometimes that’s not enough. The other day, we ran out of
everything and we had to go to a church and get food.”
6 Qualitative Data Analysis Methods
1. Content analysis
2. Narrative analysis
3. Discourse analysis
4. Thematic analysis
5. Grounded theory
6. IPA
Traditionally, researchers organized
their data by developing conceptual
files. Now, computer software
(CAQDAS) is widely used.
 Analysis of Qualitative Data

➔ Broad categories - clusters of codes that are

connected conceptually
➔ Theme - is a recurring regularity that captures
meaningful patterns in the data.
 ANALYTIC PROCEDURES
QUALITATIVE CONTENT ANALYSIS THEMATIC ANALYSIS
● Breaking down data into smaller units, ● Is another broad approach to extracting
coding and naming the units according themes from descriptive qualitative data.
to the content they represent, and ● Their step-by-step guide covers an analytic
grouping coded material based on process with six phases:
shared concepts. 1) familiarizing oneself with the data
2) gathering initial codes
● The content analysis often refers to
3) searching for themes
meaning units.
4) reviewing themes
● Manifest content: what the text usually 5) defining and naming themes
says 6) producing the report
● Latent content: what the text talks
about
 ETHNOGRAPHIC ANALYSIS
Analysis begins as the researcher enters the field. Spradley’s method
involves four levels of analysis:
1. Domain analysis - identifying domains or units of cultural knowledge
2. Taxonomic analysis - selecting key domains and constructing
taxonomies
3. Componential analysis - comparing and contrasting terms in a
domain
4. Theme analysis - to uncover cultural themes.
Interpretation of Credibility of
Quantitative Quantitative
Results Results
Aspects of Interpretation Proxies and Interpretation
Inference and Interpretation Credibility and Validity
The Interpretive Mindset Credibility and Bias
Credibility and Corroboration
02 Trustworthiness
and
Integrity in
Qualitative Research
DATA & METHOD TRIANGULATION
Triangulation uses multiple referents to draw conclusions about truth,
aiming to reduce biases from single methods, observers, or theories. It creates
a more complete understanding of a phenomenon and includes four types: data,
investigator, method, and theory triangulation.

Data triangulation validates conclusions through multiple sources,

including time (collecting data at different times), space (gathering data from
various sites), and person (collecting data from different types of people).

Method triangulation employs multiple data collection methods, such as

interviews and observations, to develop a comprehensive understanding.
DATA & METHOD TRIANGULATION

Example of person, space, and method

triangulation
● De Kok and an interdisciplinary team (2020) conducted an
ethnographic study of respectful maternity care in Malawi.
Data were collected through interviews and focus group
interviews with midwives, recently delivered mothers, and
"guardians" (the women's relatives or neighbors).
Observations were conducted in five facilities in Malawi.
Strategies Relating to Coding and
Analysis
● Investigator Triangulation
● Disconfirming Evidence and Competing Explanation
● Peer Review and Debriefing
● Inquiry Audit
- Strategies Relating to Presentation
● Thick and Contextualized Description
● Researcher Capability
Interpretation of
Qualitative Findings
The interpretative process in qualitative studies
relies on researchers' immersion and closeness to
the data, known as incubation. This process
involves understanding the meaning of qualitative
texts, identifying essential patterns, and drawing
insightful conclusions.
Interpretation of Qualitative Findings
● The Credibility of Qualitative Results

● The Meaning of Qualitative Results

● The Importance of Qualitative Results

● The Transferability of Qualitative Results

● The Application of Qualitative Results

LEARNING FROM
SYSTEMATIC
REVIEWS
 PRIMARY STUDY
● Is a specific research question using careful sampling and data collection procedures that
are spelled out in advance in a protocol. (i.e., original research investigations)

META-ANALYSIS
● is a statistical method that integrates and summarizes results from relevant publications
selected in the systematic review.

Basics of Meta-Analysis: A essence of a meta-analysis is that findings from each study are
used to compute a common index, an effective size.

Advantages of Meta-Analysis
● As an integration method: Objectivity
● Power (i.e., the probability of detecting a true relationship between variables)
 META-SYNTHESIS
● Are more than just a summaries of prior qualitative findings; they involve a discovery of
essential features of a body of findings and a transformation that yields new
interpretations.

The Noblit and Hare Approach (1988)

Meta-ethnography: This approach involves listing key themes or metaphors across studies and
then translating them into each other.

The Sandelowski and Barroso Approach (2007)

Meta-Summary: involves listing abstracted findings from the primary studies and calculating
manifest effect sizes.
META-AGGREGATION
● The approach to qualitative evidence synthesis used at the Joanna Briggs Institute (JBI),
which is more structured than a metasynthesis and relies on comprehensive searching and
systematic quality appraisals.