Académique Documents
Professionnel Documents
Culture Documents
MMM I (B)
Observational Design
Relates to the condition under which observations are to be made.
Statistical Design Concerns with the questions of how many items are to be observed (sample size) and how information and data gathered are to be analysed. Operational Design Deals with implementation part. How the techniques and procedures specified will carried out.
Concepts
Dependent and independent variable A concept which can take different quantitative values is called variable. e.g. Weight, Height, income, etc A phenomenon which can take different values within specific range even in decimal points are known as continuous variable. If one variable depends on other variable it is termed as a dependent variable. The variable which is antecedent is known as independent variable.
Extraneous variable Independent variable that are not related to the purpose of study but may not affect dependent variable are termed as extraneous variable.
Concepts
Control Minimize the influence or effect of extraneous variable. Confounded relationship When the dependent variable is not free from influence of extraneous variable then the relationship between dependent and independent variable is known as confounded relationship. Variables are said to be confounded by extraneous variable. Research hypothesis It is predictive statement that relates an independent variable with dependent variable.
concepts
Experimental and non experimental hypothesis testing research When purpose of research is to test a research hypothesis it is termed as a hypothesis testing research. Research in which independent variable is manipulated experimental Research in which independent variable is not manipulated non experimental Experimental and control groups In an experimental research when a group is exposed to usual conditions it is termed as control group. In an experimental research when a group is exposed to some novel and special conditions it is termed as experimental group.
concepts
Treatment Different conditions under which experimental and control group are put are referred as treatments.
Experiment The process of examining truth of a statistical hypothesis is known as experiment. Experiment units The pre determined plots or the blocks where different treatments are used are known as experimental group.
RESEARCH DESIGNS
Options
1) 2) 1) 2) 1) 2) 1) 2) 1) 2) 1) 2) 1) 2) 3) 1) 2) Exploratory study formal study Monitoring Communication study Experimental Ex post facto Descriptive Casual Cross sectional Longitudinal Case Statistical study Field setting Laboratory Simulation Actual routine Modified routine
Exploratory Research.
The purpose of exploratory studies is to formulate a problem for a more precise investigation or to develop hypotheses. However, an exploratory study can also be conducted to enhance the familiarity of researcher with the phenomena; he/she wishes to study some time later in a more scientific way.
Exploratory research
Exploratory studies also known as formulative research studies. Exploration studies are required when researcher lack a clear idea of problem. Through exploration researcher develop concepts more clearly, establish priorities, develop operational definitions and improve final research design.
Exploratory Research.
For instance,
A researcher might wish to find out which all factors /attributes are used in purchase decision; How the consumers are influenced by the different forms of communications? What is the possible explanation for a given marketing phenomenon (say, sales drop) and establish priorities for future research? A manufacturer faced with decreased sales might conduct an exploratory research to generate possible explanations.
CONCLUSIVE RESEARCH
Conclusive research is typically more formal and structured than exploratory research. It is based on large, representative samples, and the data obtained are subjected to quantitative analysis. The findings from this research are considered to be conclusive in nature in that they are used as input into managerial decision making.
Descriptive Research
Descriptive research studies are typically concerned with determining the frequency with which something occurs or the relationship between two variables. The descriptive research studies are typically guided by an initial hypothesis.
Descriptive Research
A study of trends in the consumption of a product with respect to such characteristics as age, sex, and geographic location would be a descriptive research study. Studies to examine if sales of a product depend on dealers/sales representatives satisfaction or advertising of the product, would be descriptive research studies.
Descriptive Research
Some descriptive research studies are conducted at single fixed time which gives a snapshot view of the problem. Other studies provide a moving description that helps researchers to follow events, people or consumer behaviour over periods of time.
CROSS-SECTIONAL DESIGN
LONGITUDINAL DESIGN
Descriptive / Diagnostic
Rigid (maximize reliability of the outcome and reduces bias)
Sampling Design
Statistical Design Observatio nal Design
Non probability
No preplanned Design for analysis Unstructured Instruments used for data collection
Probability
Pre planned design for analysis Structured and well thought and designed instruments for data collection Advanced decision regarding operational procedures
Operational Design
PREEXPERIMENT
TRUEEXPERIMENT
QUASIEXPERIMENT
Causal Research
A causal research is concerned with determining cause-effect relationships. Causal research studies typically take the form of experiments, because experiments are best suited to determine cause-effect relationships.
Causal Research
For instance, a manufacturer may be interested in ascertaining the effectiveness of different advertising appeals. One way for the company to proceed would be to use different ads in different geographic areas and investigate which ad generated the highest sales.
Causal Research
In effect, the company would perform an experiment and if designed properly, the company would be in position to conclude that one specific appeal caused the highest rate of sales.
Causal Research
Quite often a marketing manager wants to make causal inferences about the market behaviour. For example: to assess the impact of a change in retail price on the brands market share; the effect of different advertising themes upon sales of a product; the impact of a certain promotion offer; to measure the sales effectiveness of a new packaging; self space; market response to premium vis--vis economic quality of product.
Causal Research
A cause-effect relationships refers to relationships between two variables where one variable (characteristic or occurrence) X determines another variable (characteristic or occurrence) Y.
Causal Research
For example, if a researcher wishes to test a cause-effect relationships that sales promotion offer (X) causes higher sales of a product (Y) by comparing sales of out let exposed to sales promotion offer (X) with one that has not been exposed, he has to measure the two out lets with respect of Y, either during or after exposure to X.
Correlation
Cause-Effect Relationship
Causality
Correlation
Smoking Grades
S or S
Cause-Effect Relationship
CONDITIONS OF CAUSALITY
Two variables are said to be causally related :
If (1) There is an empirical correlation between them, (2) The relationship is not found to be the result of the effects of some third variable on each of the two initially observed, and (3) The cause precedes the effect in time.
Time Order
The causing event must occur either before or simultaneously with the effect.
IN -STORE SEVICE
SALES OF A DEPARTMENT STORE
Family Size
INCOME
INCOME
Pre-Experimental Designs..
E X
E = Experimental Group X = Incentive Scheme
O = Rate of Absenteeism
Group of sales persons exposed to training C = Group of sales persons not exposed to training R = Random Assignments of Sales persons O1 = Sale proceeds of group E before training X = Training O2. = Sale proceeds of group E after training O3 = Sale proceeds of group C before training O4 = Sale proceeds of group C after training D1 = Difference in sales of group E before and after the training D2 = Difference in sales of group C before and after the training ND = Net Difference in Sales
E =
O1 O3
O2 O4
O3 = O4 =
City where new advertising theme is introduced Sale proceeds of an outlet in City-E before advertisement A new advertising theme Sale proceeds of the outlet in City-E after advertisement City (appear to be similar as E) where new advertising theme is not introduced Sale proceeds of an outlet in City-C before advertisement Sale proceeds of an outlet in City-E after advertisement
Time-Series Designs.
Simple Time-Series Designs O 1 O 2O 3 O 4 O 5
O1 to O5 before X O6 to O10 after =
O6 O7 O8 O9 O10
Measurements of Dependent Variable introduction of X = Independent Variable = Measurements of Dependent Variable introduction of X
Formal
Completely randomized design Randomized block design Latin square design Factorial Design
True experimental
Pretest Posttest Control Group Design Post test only control group Design
Quasi experimental
Non equivalent Control Group Design Separate sample Pretest - Posttest Design Group Time series Design
Factorial Design
Validity in Experimentation
Whether measure accomplishes its claim. Internal validity Do the conclusions we draw about the demonstrated experimental relationship truly imply cause? External Validity Does an observed casual relationship generalize across persons, settings and times?