research methodology on cosmetics
DESCRIPTION
This presentation contains research methology approach for survey based on : Mumbai women do not use branded cosmeticsTRANSCRIPT
PROJECT ON
RESEARCH METHODOLOGY
SUBMITTED TO :
Prof . GITANJALI KAPOOR
GROUP MEMBERSPRADNYA BHALERAO C-03
RASHMIKA PATIL
C-27
PRACHI PITALE C-29
CHAITALI SWAMI C-36
PRATIKSHA RANE C-42
JAYASHREE PRABHU C-43
WHAT IS RESEARCH?
A research is an organized set of activities to
study and develop a model or procedure or
technique to find the results of a realistic problem
supported by literature and data such that its
objective/(s) is/(are) optimized and further make
recommendations or interferences for
implementations.
DA
A system of model, procedures and techniques
used to find the results of a research problems
is called a Research Methodology.
WHAT IS RESEARCH METHODOLOGY
RESEARCH PROCESS
PROBLEM DEFINITION
OBJECTIVE OF THE RESEARCH
RESEARCH DESIGN
DATA COLLECTION
PRIMARY DATA
COLLECTION
SECONDARY DATA
COLLECTION
DATA ANALYSIS
INTERPRETATION OF RESULTS
VALIDATION OF RESULTS
PAST RECORDS CURRENT FINDINGS
VALIDATION OF RESULTS
s
Z-TEST
Z-TEST
• Research studies• Normal probability distribution• Significance of mean• Comparison
• z=X-µH0
бp/ n
When standard deviation of population is not known then
we can standard deviation of the sample
бs= ∑(Xi-X)2
n-1
HYPOTHESIS TESTING OF PROPORTIONS
• Qualitative characteristic• Statistics of Attributes• Presence of an attribute-Success• Absence of an attribute-Failure
Testing the difference between proportion based on sample and the proportion for whole
population p - p z = p*q N-n n*N
• p-sample proportion• p-population proportion of success• q=1-p• n=Number of items in sample• N=Number of items in population
• n-102• N-65,80,000• Population of mumbai-14 million• 875:1000(female male ratio)• 47% females in mumbai
H0:- p=p
• H1:- p=p
• p=0.70• q=0.30• Observed sample proportion • p=87/102=0.8529
TEST STATISTIC (z)
p - p 0.8529-0.70 z = = p*q N-n 0.70*0.30* 6580000-102 n*N 6580000*102
=3.55
RESULT
• As H1 is two sided two tail test is applied for determining the rejection region at 5% level of significance
• Z=1.96• Calculated value(3.55) is more than the table
value(1.96) so we reject the null hypothesis
• There is significant difference between proportion of women using branded cosmetics in the sample and proportion of women using branded cosmetics in the population
Chi-Square Test
Research objective
Null Hypothesis (H0) : Mumbai women do not use branded cosmetics
Alternate Hypothesis (H1) : Mumbai women use branded cosmetics
E = 102 / 15 = 6.8
BRANDS OBSERVED FREQ (O)
EXPECTED FREQ (E)
O – E (O – E) 2 (O – E) 2 / E
Lakme 59 6.8 52.2 2724.84 400.71
Ponds 39 6.8 32.2 1036.84 152.47
Garnier 33 6.8 26.2 686.44 100.94
Revlon 31 6.8 24.2 585.64 86.12
Maybelline 6 6.8 -0.8 0.64 0.09
Loreal 15 6.8 8.2 67.24 9.88
Nivea 21 6.8 14.2 201.64 29.65
Ayur 5 6.8 -1.8 3.24 0.47
Himalaya 10 6.8 3.2 10.24 1.50
Oriflame 9 6.8 2.2 4.84 0.71
Elle 18 22 6.8 15.2 231.04 33.97
Fair & Lovely 9 6.8 2.2 4.84 0.71
Fairone 1 6.8 -5.8 33.64 4.94
Fairever 2 6.8 -4.8 23.04 3.38
Others 14 6.8 7.2 51.84 7.62
Total 833.16
∑(O-E)2 / E = 833.16
Degree of Freedom = 15-1 = 14
Assuming 5% level of significance.
T0.05,14 = 23.685
Tcalc > Tα
Null Hypothesis is rejected
Therefore, Mumbai women use branded cosmetics
Limitations Of Research
• Small population size
• Small sample size
• Only women cosmetics were considered
• Only few brands were considered
• Misunderstanding of the question can all contribute to inaccuracies in the data.
• Survey question answer-choices could lead to vague data sets
• Bias Issues
• External Environment
• Time consuming
CONCLUSION
Bibliography & Webliography
www.google.com Research Methodology &Techniques – by C.R Kothariwww.wikipedia.com Statistical Methods – by S.P. Gupta www.esurveyspro.com Statistics for Management – Richard Levin