PPTs for teaching learning for course BUSINESS RESEARCH
PPTs for teaching & learning for course BUSINESS RESEARCH METHODS MBA Sem I Savitribai Phule Pune University, Pune, India. By Prof. Swapnil Undale & Dr. Padmavati Undale This work is licensed under the Creative Commons Attribution-Share. Alike 4. 0 International License. You are free to use, distribute and modify it, including for commercial purposes, provided you acknowledge the source and share-alike. To view a copy of this license, visit http: //creativecommons. org/licenses/by-sa/4. 0/.
Experimental or Causal Research Design Types of research design- 3. Experimental research design-Studies the cause and effect relationship between variables under study. -manipulate some variable in setting and observe how it affects the participants or subjects being studied.
CAUSAL OR EXPERIMENTAL RESEARCH DESIGN An experiment is a study involving intervention by the researcher beyond that required for measurement. The usual intervention is to manipulate some variable in setting and observe how it affects the participants or subjects being studied. (People or physical entities) The researcher manipulates the independent or explanatory variable and then observes whether the hypothesized dependent variable is affected by the intervention.
Causality: A concern with establishing causal connections between variables, rather than mere relationships between them e. g. A B C Z C D E Z there fore C Z A B C Z A B No C No Z therefore C Z Why sale is not growing? Poor promotion correct increased sale Poor implementation correct increased sale Poor sales call correct increased sale
Generalization: Findings should be generalizable Findings should be applicable to large population e. g. Why sales degrowing? Cannot study every individual HQ. Select samples HQs randomly Analyze ? Promotion ? Implementation ? Sales call We should be able to say confidently that result of this study is applicable to all HQs across the country.
Replication: Process of repeating the experiment to establish its reliability The degree to which the result of a study can be reproduced e. g. Theory of gravitation An apple falls on the earth A mango falls on the earth A ball falls of the earth ? Sales degrowth Gr. 1 Poor implementation Repeat Gr. n Poor implementation
Variables Variable : A concept which can take on different quantitative values. ( weight, height, income, satisfaction etc. )
Types of variables – Continuous – The variable which can assume any numerical values within a specific range. The variable which can take on quantitatively different values even in decimal points e. g. income, age, temperature, etc. Discrete/Non-continuous. The variable which can only expressed in integer values. A variable which individual values fall on scale only with distinct gaps. E. g. Students, Children, vehicles owned Dichotomous : A variable for which only two options are Available. <2 B/not 2 B> e. g. alive or dead, graduate or non-graduate.
Types of variables – Dependent Variable (DV)/ criterion variable: A measured, predicted or otherwise monitored variable expected to be affected by manipulation of an IV. -Can not be manipulated by researcher Independent Variable (IV)/Explanatory/Predictor variable: The variable which can be manipulated by the researcher, thereby causing an effect on the DV. e. g. Advertisement & Sales, Incentive & productivity.
Types of variables – Extraneous : The variables which are not studied but affect the relationship of the studied variables. Most of them can be ignored or their effect on the relationship is insignificant. e. g. Gender Male or Female Intervening : The variable which theoretically affects the observed phenomenon but cannot be seen, measured or manipulated. Its effect can be inferred from the effect of independent variable on the dependent one. E. g. Motivation, Ego
Types of variables – Moderating : A second variable that is included in the relationship study because it is believed to have a significant contributory effect on the IV – DV relationship. e. g. Age (MV) of the employee will significantly affect the incentive (IV) and productivity (DV) relationship. Gender (EV) Gr 1(M) Gr 2(F) Gr 3(M&F) Degree of Motivation (Intervening) to earn incentive
Important Terms in experimentation CONTROL GROUP (CG) A group that is measured but not exposed to experimental treatment. Treatment Group (TG) A group that is measured and exposed to experimental treatment. Randomization (R) Random assignment of subjects to control and treatment group and random allocation of treatment.
Formal Experimental designs √ Completely randomized design √ Randomized Block Design √ Latin Square Design √ Factorial Design
Formal Experimental designs 1. Completely Randomized Design A. Two group Simple Randomized design Ø Required samples are selected from the population under study Ø Two groups are formed 1. Control group (CG) 2. Experimental group (EG) Ø Samples are randomly assigned to these two groups Ø EG is exposed to treatment (X) Ø Both groups are measured/observed (O 1 & O 2) Ø difference between both group is effect of treatment
Formal Experimental designs 1. Completely Randomized Design A. Two group Simple Randomized design (Contd…) Ø e. g. setting up design to measure effect of new teaching method Ø Population = Students Ø Sampling = Random Ø Two groups = EG & CG Ø Two teaching method = Old (X 1) & New (X 2) Groups Pre-treatment Treatment Post - treatment CG (R) O 1 X 1 O 2 EG (R) O 3 X 2 O 4 Ø Effect of treatment (New method) = (O 4 -O 2) – (O 3 -O 1)
Formal Experimental designs 1. Completely Randomized Design B. Random replication design Ø Required samples are selected from the population under study Ø Required treatments are selected randomly Ø More than 2 groups are formed 1. Control group (CG 1. CG 2, CG 3……. CGn) 2. Experimental group (EG 1, EG 2, EG 3……EGn) 3. Treatments (X 1, X 2, X 3, ……. . Xn) Ø Samples are randomly assigned to these groups Ø EG is exposed to treatment (X) to EG is selected randomly. Ø All groups are measured/observed (O) Ø Average difference between the groups is effect of treatment
Formal Experimental designs 2. Randomized Block Design Based on certain pre-decided criterion, homogeneous groups are formed Rest procedure same as C. R. design. e. g. students are divided as per their core domain (EV) (Arts, Commerce & Science) Groups Pre test Treatment Post test CG (R) A O 1 X 1 O 2 EG (R) A O 3 X 2 O 4 CG (R) C O 5 X 1 O 6 EG (R) C O 7 X 2 O 8 CG (R) S O 9 X 1 12 EG (R) S 13 X 2 14
Formal Experimental designs 3. Latin Square Design - study the effect two independent variable. e. g. Effect of teaching method depend upon core knowledge & IQ level Groups IQ level Pre-treatment Treatment Post - treatment CG (R) A Bellow Avg O 1 X 1 O 2 Above Avg O 3 X 1 O 4 Bellow Avg O 5 X 2 O 6 Above Avg O 7 X 2 O 8 Bellow Avg O 9 X 1 O 10 Above Avg O 11 X 1 O 12 Bellow Avg O 13 X 2 O 14 Above Avg O 15 X 2 O 16 Bellow Avg O 17 X 1 O 18 EG (R) A CG (R) C EG (R) C CG (R) S
Formal Experimental designs 3. Latin Square Design - study Groupsthe effect IQ leveltwo independent Pre-treatment variable. Treatment Post - treatment e. g. Effect depend & IQ level CG (R)of A teaching Bellow method Avg O 1 upon core knowledge X 1 O 2 EG (R) A CG (R) C EG (R) C CG (R) S EG (R) S Above Avg O 3 X 1 O 4 Bellow Avg O 5 X 2 O 6 Above Avg O 7 X 2 O 8 Bellow Avg O 9 X 1 O 10 Above Avg O 11 X 1 O 12 Bellow Avg O 13 X 2 O 14 Above Avg O 15 X 2 O 16 Bellow Avg O 17 X 1 O 18 Above Avg O 19 X 1 O 20 Bellow Avg O 21 X 1 O 22 Above Avg O 23 X 2 O 24
Formal Experimental designs 3. Latin Square Design : The experiment can be graphically shown as below. Age → Investor ↓ Young Middle aged Old Small R O 1 X 1 O 2 R O 3 X 2 O 4 Medium R O 7 X 2 O 8 R O 9 X 3 O 10 R O 11 X 1 O 12 Large R O 13 X 3 O 15 X 1 O 16 R O 17 X 2 O 18 O 14 R R O 5 X 3 O 6
Formal Experimental designs Factorial Design : Two Treatments – X = Interest Rate and Y = Deposit amount Investors are grouped according to type of investor and age of investor. The experiment can be graphically shown as below. Age → Investor ↓ Young Middle aged Old Small R O 1 X 1 Y 1 O 2 R O 3 X 1 Y 2 O 4 R O 5 X 1 Y 3 O 6 Large R O 7 X 2 Y 1 O 8 R O 9 X 2 Y 2 O 10 R O 11 X 2 Y 3 O 12 Effect of each of the combination of treatments will be studied for two variables like small and young or Large and middle aged and so on.
THANK YOU
- Slides: 22