Statistics in nursing research

Descriptive statistics offer nurse researchers valuable options for analysing and pre-senting large and complex sets of data, suggests Christine Hallett

1. STATISTICS IN
NURSING
RESEARCH
1

2. ▫Statisticsis the science of making effective use of numericaldata
which is related to collection, analysis and interpretation of data.
Statistics is the study of how to collect, organizes, analyze, and
Interpret data.
DEFINITION
IMPORTANCE
•Statistics plays a vitally important role in the research.
•It help to answer important research questions and field in study.
•Helps you understand how to apply statistical method
•Important to understand what tools are suitable for a particular
research study.
•Statistics enables to understand specified statistical concepts
and procedures.
2
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

3. TYPES OF STATISTICS
Therearetwoapproachestothestatisticalanalysisofdata
1.DescriptiveStatistics
Descriptivestatisticsaretechniqueswhichhelptheinvestigator
toorganize,summarizeanddescribemeasuresofasample.
2.Inferentialstatistics
Theinferentialapproachhelpstodecidewhethertheoutcome
ofthestudyisaresultoffactorsplannedwithindesignofthe
studyordeterminedbychance.
(Streiner&Norman,1996).
3
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

4. FREQUENCY DISTRIBUTIONS
Frequency distribution is a systematic
arrangement of values from lowest to highest or a
method of organizing numeric data
22 23 25
23 16 20
15 24 23
24 23 16
23 18 22
20 25 25
No.s (x)Frequency
(f)
15 l
16 ll
18 l
20 ll
22 ll
23 llll
24 ll
25 lll
18Σf =
4
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

5. PRESENTATION OF DATA & SHAPES
▫Tabular presentation
▫Diagrammatic Presentation
▫Graphical Presentation
A. Tabular Presentation of Data
▫Arranging values in columns is called tabulation.
▫E.g.The amount of oxygen content in water samples
B. Diagrammatic Presentation of data
▫It is a visual form of presentation of statistical data in which data are
presented in the form of diagrams such as bars, lines, circles, maps
▫Common Types
▫Line Diagram
▫Pie diagram
▫Bar diagram
c. Line diagram
5
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

6. SHAPES OFFREQUENCY DISTRIBUTION
2. Polygons: polygons
use dots connected by straight lines
to show frequencies.
1. Histograms: A histogram is
constructed by drawing bars
Distribution are shown in Graphically. Graphs denotes the
information of complete data in different shapes
6
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

7. Conti..
3. Symmetric distribution (Normal )
It consist of two halves that are mirror images of
one another.
4. Asymmetric or Skewed distribution
It is off center and one tail is longer than the other
If the tail points to the left, the distribution
is negatively skewed,
-When the longer tail points to the right,
the distribution is positively skewed.
A distribution with the modal peak off to one side or the other is
described as skewed. The word skewliterally means "slanted."
7
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

8. Conti..
5. Unimodal distribution
It has only one peak or high point
•(i.e., a value with small / high frequency),
6. Multimodal distribution
It has two or more peaks
(i.e., values of high frequency).
8
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

9. STATISTICS & DATA ANALYSIS
Measures of central tendency
Mean
Mode
Median
Measures of variability
Range
Standard deviation
Correlation
Inferential statistics
T-test
Chi square test
ANOVA
9
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

10. CENTRAL TENDENCY
It is a statistical measure that identifies a single scor
e as representative for an entire distribution or group.
1.Mean
2.Mode
3.Median
Levels of measure used:
Interval level variables -Mean
Nominal variables -Mode
Ordinal variables -Median
Measures of Central Tendency
10
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

11. 1. Mean
•The mean (or average) is adding all the numbers and then
divided by the number of observations in the data set.
Example: 3,4,5,6,7
▫3+4+5+6+7= 25, 25 N=5 The mean = 5
Exercise 1.
What is the average of these numbers?
567, 432, 902, 693, 356, 996
11
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

12. 2. Mode
The mode in a set of data is the number that occurs
the most. E.g 25 10 10 25 5 1025 10 5
Exercise 2: Find the mode of these numbers.
100 95 100 90 75 100 85 95
3. Median
The median is a set of data , which is the middle
number. Also arrange all the data from lowest to highest and
then take the middle number.
E.g : odd : 3, 5, 8, 10, 11 median=8
even: 3, 3, 4, 5, 7, 8 median=(4+5)/2=4.5
Exercise 3: Find the median
1. 67 34 85 33 84 & 2. 12 14 16 18
19 20
12
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

13. Relationship between mean, median, and mode is
determined by the shape of the distribution
CENTRAL TENDENCY AND THE SHAPE OF THE
DISTRIBUTION
13
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

14. Variability provides a quantitative measure of the degree t
o which scores in a distribution are spread out or clustered
together.
If data has two distributions (Bivariante) with the same
mean known as variability and have different shapes.
VARIABILITY (Disperson)
Measure of variability or Disperson
Range
Standard deviation
Correlation & co-efficient
14
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

15. Range
It is the difference between the lowest and highest
number in the set.
Range = X
highest –X
lowest
E.g: SAT scores of students at two nursing schools. Both distributions
have a mean of 500, but the score patterns are different. School A has a
wide range of scores, with some below 300 and some above 700. This
school has many students who performed among the best also many
students who scored well below average. In school B, there are few
students at either extreme.
15
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

16. STANDARD DEVIATION
Standard deviation is the most common measure of variability.
It is used the mean as a reference point and approximates the
average distance of each score from the mean.
•A deviation (x) is the difference between an individual score
and the mean.
•VARIANCE
▫The variance is simply the value of the standard
deviation before a square root has been taken
16
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

17. Formulas of standard Deviation
Standard Deviation (s or SD)* is
-1
st
column is mean (X) = 7
-2
nd
column = (X -X )
-3
rd
column each score is squared
( X –X )2
.
17
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

18. CORRELATION
Correlation is a measure of association between two
variables. Correlations can be graphed on scatter plot or
scatter diagram
Scatter plot: It involves making a rectangular
coordinate graph with the two variables laid out at right
angles. plot (dots) are shown to help identify subjects.
18
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

19. Positive Correlations : If the dots begins at the lower left corner and
extends to the upper right corner, the relationship is positive .
Correlationsfall between .00 and 1.00 are positive
Negative Correlations: Dots from the upper left corner to the lower
right corner, the relationship is Negative. Correlations that fall
between .00 and 1.00 are negative,
19
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

20. CORRELATION COEFFICIENT (PEARSON’S –R)
Correlation coefficients can be computed with
two variables measured on either the ordinal, interval,
or ratio scale
Pearson’s
Calculation…..
20
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

21. INFERENTIAL STATISTICS
•Inferential statistics is a statistical method used to infer result s
of sample (statistic) to population (parameter).
It is a process of inductive reasoning based on the mathematical
theory of probability
-(Fowler, J., Jarvis, P. -2002).
•Component of inferential statistics.
▫Hypothesis testing
▫Estimation
21
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

22. Error and Hypothesis testing
The standard deviation of a sampling distribution of meanis
called the Standard Error of the Mean (SEM).
Error
various means in the sampling distribution have some error as
estimates of the population mean
SEM (symbolized as Sx)
If we use this formula to calculate the SEM for an SD of 100
with a sample of 25 students we obtain
22
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

23. Errors
Reject H0 Don't reject H0
Truth
H0Type I ErrorRight decision
H1Right decisionType II Error
Type I error ()
Accepting the experimental hypothesis when the null hypothesis
is true
Type II error ()
Accepting the null hypothesis when the experimental hypothesis
is true
Conti…..
23
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

24. Hypothesis testing
•A study was conducted to determine the difference of knowledge score of
hypertension between male and female adults in Bandar A. the result
revealed t statistic 2.678, df 99, P value was 0.009(level of significance set at
0.05) and mean difference 1.14
•Hypothesis
▫H
O: there is no difference of knowledge score of hypertension
between male and female adults in Bandar A
▫H
A: there is the differences of knowledge score of hypertension between
male and female adults in Bandar A
•Interpret the result
▫P=0.009, α=0.05, p<α. Reject H
O
•conclusion
▫There is difference of knowledge score of hypertension between male and
female adults in Bandar A
24
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

25. ESTIMATION
It is used to estimate a single parameter, like a mean.
Estimation can take in to two forms.
Forms:
Point estimation : Point estimation involves calculating a
single statistic to estimate the population parameter.
Point estimates convey no information about accuracy
Interval estimation : it indicates a range of values within
which the parameter has a specified probability of lying
25
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

26. STATISTICAL TESTS
There are two types of inferential statistics
1.Parametric
2.Non-parametric Tests
1. Parametric Tests
A parametric test is one which specifies certain conditions
about the parameter of the population from which a sample is taken.
E.g t-test,andF-test (ANOVA)
2. Non-parametric tests (Distribution-free Statistics)
A non-parametric test is one does not specify any conditions
about the parameter of the population from which the population is
drawn. These tests are called.
E.g Chi-squire test
26
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

27. T-TEST (Student’s t).
•It is used to testing the differences in group s of mean
•t-test can be used when there are two independent groups
(e.g., experimental versus control, male versus female),
Degree of freedom (df)
•Degree of freedom (df) is describes the number of events or observations
that are free to vary.
Formula
t-Test Degrees of freedom (df)
27
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

28. Example:
Maternity patients on perceived maternal competence. We
administer a scale of perceived maternal competence at
discharge to 10 primiparas who remained in the hospital 48 to
72 hours (group A, regular discharge) and to 10 primiparas
discharged less than 48 hours after delivery (group B, early
discharge). The mean scale scores for these two groups are
25.0 and 19.0,
nA = 10 number of subjects in group A
nB =10 number of subjects in group B
28
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

29. Degrees of freedom :
df = [(10 + 10) -2] = 18
29
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

30. THE CHI-SQUARE TEST (Analyzing Frequencies)
The chi-squire test is used when the data are expressed in
terms of
frequencies of proportions or percentages.
The chi-square statistic is computed by comparing observed
frequencies and expected frequencies
FORMULAS
Chi-square
Degrees of freedom (df) = [(R 1)(C 1)].
Calculation…….
30
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

31. Analysis of variance (ANOVA)
It is another commonly used parametric procedure for testing
differences between means where there are three or more
groups.
The statistic computed in an ANOVA is the F-ratio , variation
within groups to get an F-ratio.
Types
One way Anova, two way ANOVA, multifactor ANOVA
Formulas
Calculation…..
MEAN SQUARE (MS)
F-Ratio
31
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

32. TYPES of ANOVA
•One-way ANOVA
▫Itis used with one independent variable and one dependent variable).
•Two-way ANOVA or Factorial Analysis of Variance
▫Factorial analysis of variance permits the investigator to analyze the
effects of two or more independent variables on the dependent
variable.
•Analysis of Covariance (ANCOVA)
▫It is an inferential statistical test that enables investigators t adjusts
statistically for group differences that may interfere with obtaining
results that relate specifically to the effects of the independent
variable(s) on the dependent variable(s).
•Multivariate Analysis
▫Multivariate analysis refers to a group of inferential statistical tests that
enable the investigator to examine multiple variables simultaneously.
32
Nursing Path
6/22/2016www.drjayeshpatidar.blogspot.com

33. Thank You
33
Nursing Path