NURSING RESEARCH 2 (BIOSTATISTICS

INTEGRATED) | NCMB 315

| Midterms | 2nd Semester, 3rd Year
QUANTIFYING RESEARCH DATA What We Need to Know When Quantifying
Phase 4: The Analytic Phase Data
o Analyzing through appropriate quantitative or o Understand the research problem
qualitative methods, the research data o Understand the nature of our variables
o Interpreting the results of the analysis. ➢ Categorical or metric data
➢ Levels of measurement: noir
Statistics o Level of Statistical Analysis
o Deals with logical collection, organization, o Distinguish the appropriate statistical procedures
presentation, analysis and interpretation of numerical
data. Nature of Research Problem
o Do we intend to describe? Infer from the sample?
Fields of Statistics Explore the data?
o Descriptive: allow the researcher to organize the o Are we concerned with relationships? Differences?
data in ways that give meaning and facilitate insight o Are we interested in causal relationships? Predictive
(frequency distributions and measures of central relationships?
tendency and dispersion)
o Inferential: statistics designed to allow inference Nature of Variables
from a sample statistic to population parameter; o Categorical Data
commonly used to test hypotheses of similarities and ➢ Cases are in defined classes; data are counted
differences in subsets of the sample under study or yield frequencies
❖ e.g. Gender (variable), categories–male
Statistics and its Place in Research and female
o The Research Process o Metric Data
➢ Conceptualization → Choice of Research ➢ Cases are measured. Thus, these data yield
Method → Population and Sampling → metric or scale values.
Observations → Data Processing → Data ❖ e.g. height (4.5 meters), IQ (130)
Analysis → Application
Levels of Measurement
Statistics and its Importance in Research o Nominal: Data are classified into categories. These
o A means to achieve the goals of science: categories have no particular order.
➢ Gives exact kind of description ➢ Gender, civil status, political affiliation, etc
➢ Enables us to draw conclusions o Ratio: highest level of measurement. Aside from a
➢ Enables us to predict constant size in difference between numbers, it has a
➢ Helps us analyze causal relationships fixed zero point.
o Reinforces systematic procedures ➢ Weight, height, income, allowance per week
o Brings order out of chaos o Ordinal: Categories imply some sort of ranking
➢ Year Standing, Ranks of Professors, Likert scale,
The Uses of Statistics
etc., SES
o According to records, a Russian lady had sixty- nine
o Interval: the difference between numbers is a known
children: sixteen sets of triplets and four sets of
constant size; zero is arbitrary
quadruplets.
➢ Temperature in Celsius, IQ scores
➢ Statistics can be employed to summarize data
and describe them economically: descriptive
statistics
o As inferred from the sample, it can be deduced that
females are better than males when it comes to math
➢ More commonly used in research, statistics is a
tool to test hypotheses: inferential statistics
o Based on existing records, an average male smoker
started smoking at 15, while an average female
smoker started smoking at 18. Thus...
➢ Statistics can be used to generate hypotheses:
exploratory data analysis

Nicole Navarro | 1
NURSING RESEARCH 2 (BIOSTATISTICS
INTEGRATED) | NCMB 315
| Midterms | 2nd Semester, 3rd Year
Levels of Statistical Analysis ❖ The information gathered could guide
o Univariate Analysis – examination of the distribution business planning and decision making.
of cases on only one variable.
o Bivariate Analysis – two variables are studied; an Measures of Dispersion
o Applicable to at scales
element of comparison is added.
o Range: highest value – lowest value
o Multivariate Analysis – several variables are being
o Variance and Standard Deviation: variance is the
studied
average of the squares of the distance each value is
Mean, Median and Mode from the mean (Bluman, 1993:95).
o Mean – average, add all values divided by the ➢ The square root of the variance is the standard
number of values deviation.
o Median – the value which divides the values into two o Standard Deviation: how disperse the values are
halves from the mean
o Mode – most frequent occurring value
Choosing The Appropriate Measure
Descriptive Statistics Univariate Analysis: Organizing Data
o It is a statistical technique that provides simple o Nominal
description and summary about the sample and ➢ Tabular Presentations:
about the observations that has been made. ❖ Frequency, Percentage, Proportion
Distribution, Rates, Ratio Cumulative
Choosing The Appropriate Measure Frequency
Univariate Analysis: Describing Data ➢ Graphical Presentations:
Measures of Central Tendency ❖ Bar chart, Line chart, Paretograph, Pie chart
o Mean Maps, pictographs
➢ Level of Measurement: Interval and Ratio o Ordinal
➢ Purpose: ➢ Tabular Presentations:
❖ Best measure to use when every single ❖ Frequency, Percentage, Proportion
observation is needed to contribute to the ❖ Distribution
average. ❖ Cumulative Frequency
❖ Usually, normal distributions ❖ (arranged in order)
o Median ➢ Graphical Presentations:
➢ Level of Measurement: Ordinal (interval and ❖ Bar chart, Line chart, Paretograph, Pie
ratio) chart, Maps
➢ Purpose: ❖ (Arranged in Order)
❖ Estimation of the cost of an average o Scale
housing unit in a community (since the ➢ Tabular Presentations:
median is not affected by extreme scores, ❖ Ungrouped or Grouped frequency
it would probably give a more accurate (percentage or proportion) distribution;
picture of the cost of a typical housing arranged on the basis of scores
unit) ➢ Graphical Presentations:
❖ Other examples: economic level of a ❖ Histogram
particular community (distribution of ❖ Frequency polygons
incomes), marrying age, etc. ❖ Ogives
❖ Usually for skewed distributions ❖ Stem and leaf display
o Mode
➢ Level of Measurement: Nominal (O, I, R) Choosing The Appropriate Measure
➢ Purpose: Bivariate Analysis: Describing Data
❖ Has its uses and advantages especially for Nominal
producers, designers, storekeepers, o Phi coefficient
salesmen, businessmen, who aim their ➢ Or mean square contingency coefficient and is
products at specific markets. a measure of association for two binary
❖ Ex: most frequently bought item or variables. Introduced by Karl Pearson, this
knowing what price is most frequently paid measure is similar to the Pearson correlation
for certain commodities coefficient in its interpretation.
Nicole Navarro | 2
NURSING RESEARCH 2 (BIOSTATISTICS
INTEGRATED) | NCMB 315
| Midterms | 2nd Semester, 3rd Year
Ordinal Inferential Statistics
o Spearman Rho (Maximum value is 1) Measures of Association
➢ is a nonparametric measure of rank correlation o Level Of Measurement
(statistical dependence between the ranking of ➢ Nominal
two variables). ❖ Correlation Coefficients:
➢ Spearman's rank correlation coefficient or • Phi Coefficient (2x2)
Spearman's rho, named after Charles • Contingency coefficient
Spearman • Cramer’s Statistic
➢ is the nonparametric version of the Pearson ➢ Ordinal
correlation coefficient. Your data must be ❖ Correlation Coefficients:
ordinal, interval or ratio. Spearman’s returns a • Spearman Rho
value from-1 to 1, where: • Kendall’s Tau
❖ +1 = a perfect positive correlation • Kruskall-Wallis
between ranks ➢ Scale
❖ -1 = a perfect negative correlation ❖ Correlation Coefficients:
between ranks • Pearson R
❖ 0 = no correlation between ranks.
o Kruskal-Wallis H test
➢ Sometimes also called the "one-way ANOVA
on ranks"
➢ Is a rank-based nonparametric test that can be
used to determine if there are statistically
significant differences between two or more
groups of an independent variable on a Type of Research Statistics
continuous or ordinal dependent variable. Question
➢ The test determines whether the medians of Descriptive Mean, frequency
two or more groups are different. Like most Complex Descriptive Cross-tabulations, factor
analysis
statistical tests, you calculate a test statistic and
Single factor difference T-test, Z-test, one-way
compare it to a distribution cut-off point.
questions ANOVA
Complex or multifactor Factorial ANOVA
Scale difference questions
o Pearson R Basic associational Correlation
➢ is a statistical formula that measures the questions
strength between variables and relationships. Complex or multivariate PearsonR; when
❖ To determine how strong the relationship associational questions predicting: Multiple
is between two variables, you need to find Regression
the coefficient value, which can range
between-1.00 and 1.00.
Choosing the Appropriate Measure
➢ If the coefficient value is in the negative range,
then that means the relationship between the
Inferential Statistics
o For 2 (or more than) independent variables, use
variables is negatively correlated, or as one
factorial ANOVA
value increases, the other decreases.
➢ this test enables us to understand the separate
❖ If the value is in the positive range, then
(main effect) influences of each IV and their
that means the relationship between the
interactional effects on a certain DV.
variables is positively correlated, or both
o When correlating more than 2 variables, use
values increase or decrease together.
Pearson R to calculate the multiple correlation
➢ p- value = Alternative approach in decision
coefficient (make sure your data are at least interval).
making
o When predicting, use regression analysis (your data
➢ Decision rule: if p <a (0.01 or0.05) Reject the H₀
must be at least interval)
conclude Ha

Nicole Navarro | 3
NURSING RESEARCH 2 (BIOSTATISTICS
INTEGRATED) | NCMB 315
| Midterms | 2nd Semester, 3rd Year
Analyzing Data o Type II
o Level of significance: α, probability of rejecting a ➢ Error that occurs when the researcher concludes
true Ho hypothesis that no significant difference exists between the
o Setting the level of significance is setting the samples examined when, in fact, a difference
probability of erroneously rejecting a true Ho to be at exists; the null hypothesis is regarded as true
the most equal to α when it is false
o α is conventionally set at 0.05, 0.01 or 0.1 e.g. ➢ Type II: (False Negative Error)
rejecting Ho set at α = .05 means that the probability ❖ A type II error occurs when the null
of rejecting a true Ho is at most only 5%. Or 5 times hypothesis is false but was accepted as
in 100 samples true by the testing.
❖ A type II error, or false negative, is where a
Interpreting the Results test result indicates that a condition failed,
Hypothesis Testing while it actually was successful.
o Statistical Hypothesis is subjected to statistics. ❖ The person is sick. But after the test he or
o CV =/< TV ---- accept the Ho she is healthy or not sick.
o CV > TV ------ reject the Ho
o ERRORS
➢ Type I
➢ Type II

Errors
o Type I
➢ Error that occurs when the researcher concludes
that the samples tested are from different
populations (a significant difference exists
between groups) when, in fact, the samples are
from the same population (no significant
difference exists between groups); null
hypothesis is rejected when it is true
➢ Type I: (False Positive Error)
❖ A Type I error is the incorrect rejection of a
true null hypothesis.
❖ Is asserting something as true when it is
false. This false positive error is basically a
“false alarm”
❖ In reality: a person is not sick. But after the
test the result is (he or she is sick)

Nicole Navarro | 4