4 Research Methods in Managem The Business Research
4 Research Methods in Managem The Business Research Process: An Overview Causality in research © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Causal Research • Research conducted to identify cause and effect relationships (inferences). Cause Effect © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
B A (1) A causes B (Unidirectional relationship) A B (2) B causes A (Unidirectional relationship) B A (3) A and B cause each other (Bidirectional relationship) B A (4) A and B just correlate (The direction of causality cannot be implied) © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Causal Research • Unidirectional relationship example Heavy rain Car accident Heavy rain causes car accident Heavy rain X Car accident However, car accident cannot cause heavy rain © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Causal Research • Bidirectional relationship example The more we eat, the more we get fat Eating Gaining fat When we get fat, the more we eat The more we exercise, we become stronger Exercise Getting strong When we are strong, we can exercise more © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Evidence of causality To prove that a causality is real, there are 3 conditions that must be met. • Temporal sequence • Concomitant variation • Nonspurious association © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 4– 6
Evidence of causality • Temporal sequence—the appropriate causal order of events. Cause Effect Happened at Time t (e. g. year 2000) Happened at Time t+1 (e. g. year 2001) Cause Effect Happened at Time t-1 (e. g. year 2001) Happened at Time t (e. g. year 2000) © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 4– 7
Evidence of causality • Temporal sequence—the appropriate causal order of events. • Some event takes time to bring about the effect ◗ ◗ Eating (today) → Gaining weight (next week) Unhealthy lifestyle (this year) → Cancer (20 years later) Investment (this year) → Profit (12 years later) Training (this month) → Skill improvement (3 months later) • Temporal sequence can be tested by using “Time-lag data” to analyze the causal relationship • The data of “Cause” variable is collected at time t • The data of “Effect” variable is collected at time t+n © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Evidence of causality • Concomitant variation—two phenomena vary together. • Occur when two events “covary” or “correlate” • Change in Y must follow the change in X X increases when Y increased X decreases when Y increased © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 4– 9
Evidence of causality • Concomitant variation—two phenomena vary together. X Hours spent on study Y Test score 1 3 2 4 5 6 6 7 7 9 8 8 8 Y 7 6 5 4 3 2 1 1 2 3 4 5 6 7 8 © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. X 4– 10
Evidence of causality • Nonspurious association—an absence of alternative plausible explanations. • For example: • There is a positive relationship between “ice cream sales” and “murder cases”. • When ice cream sales increase, murder cases increase. • When ice cream sales drop, murder cases reduce. © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 4– 11
EXHIBIT 4. 2 The Spurious Effect of Ice Cream © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 4– 12
Evidence of causality • Nonspurious association—an absence of alternative plausible explanations. When analyzing the relationship between variables, it is essential to take into consideration other factors that might also influence the dependent variable. Cause Effect Control variables Hidden factor 1 Hidden factor 2 These factors are called “control variable” © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 4– 13
Association does not imply causation • a correlation between two variables does not necessarily imply that one causes the other. © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 14
Examples of Spurious Association © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 15
Examples of Spurious Association © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 16
Examples of Spurious Association © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 17
Reverse causation • Reverse causality means that the direction of causality between two factors may be opposite from what we expect ? OR ? © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 18
Reverse causation • Example: Considered the relationship between “Income” and “Health”. . which one is the cause? . . which one is the effect? Having money allows people to gain access to better healthcare service Income Health Having good health allows people to be more productive to earn more money © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 19
Reverse causation Example: Considered the relationship between “the amount of bank loan” and “business growth”. . which one is the cause? . . which one is the effect? Companies need loan to grow their business Loan amount Business growth As business is growing, companies can obtain higher amount of loan from the bank © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 20
Different types of effect Causality analysis in academic research can be presented in three pattern • Main effect • Moderating (interacting) effect • Mediating effect © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12– 21
Main Effect The influence of a single independent variable on a dependent variable. Independent Variable (X) Dependent Variable (Y) Main effect © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12– 22
Moderating (Interaction) effect • Moderating effect • The effect of the independent variable (cause) on the dependent variable (effect) is contingent on the third factor which can either enhance or suppress the effect. • The third factor that influences the relationship is called a “moderator”. © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12– 23
Moderating (Interaction) effect Moderator Moderating effect Independent Variable (X) Dependent Variable (Y) Main effect © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Moderating (Interaction) effect Does a training program increase sale performance? Number of training sessions Sale performance © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Moderating (Interaction) effect Young employees Old employees © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Moderating (Interaction) effect Young employees Old employees © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12– 27
Moderating (Interaction) effect • Age of employee “moderates” the effect of training program on sale performance. • Only “young employees” who “received more training” tend to have higher sale performance Age of employee Number of training sessions Sale performance © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Does using social media at work affect job performance? Social media use at work Job performance © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 1– 29
© 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 1– 30
Do employees who have higher job stress benefit more from using social media during work than employees who have lower job stress? Job stress Social media use at work Job performance © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 1– 31
© 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 1– 32
Do employees who access to social media from PC benefit from using social media during work more than employees who access from mobile devices? Access from PC or Mobile devices Social media use at work Job performance © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 1– 33
© 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 1– 34
Do employees who work for a firm that allow social media at work obtain more benefit than employee who work for a firm that prohibit social media at work? Social media are prohibited at work Social media use at work Job performance © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 1– 35
© 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 1– 36
Social media is prohibited at work Access from PC or Mobile gadgets Social media use at work Job performance Job stress © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 1– 37
Mediating effect • X does not influence Y directly. • But it can influence Y indirectly through a third variable called a “Mediator” • M mediates the relationship between X and Y Mediator Independent Variable (X) Dependent Variable (Y) © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Mediating effect • Full mediation • occurs when the direct relationship between the independent variable and the dependent variable disappear when the mediator is included. • Partial mediation • occurs when the direct relationship between the independent variable and the dependent variable is still strong even when the mediator is included. © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Strong relationship Independent Variable (X) Mediator Strong relationship Weak relationship or no relationship Dependent Variable (Y) 1– 40 © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Strong relationship Independent Variable (X) Mediator Strong relationship Dependent Variable (Y) 1– 41 © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Example of mediating effect Why does heavy rain lead to high death rate? Does heavy rain directly kill people? NO!! Heavy Rain Weak relationship Death rate © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Example of mediating effect Heavy rain does not directly cause death. However, heavy rain increases the chance of car accident, which in turn, increases death rate. Po sit ive an ds tro ng Chance of Car accident Heavy Rain Weak relationship ive an d str on g Death rate High chance of car accident “fully mediates” the positive linkage between Heavy rain and Death rate © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Example of mediating effect Research found the effect of weather condition and stock prices Effect © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Example of mediating effect Weather condition does not directly affect stock price. However, weather condition affect moods of investors, which in turn, affect their decision to buy/sell stocks. Moods of investors t St Str on ge ffe c ro ng Weather condition Weak effect ef fe c t Stock market movement Moods of investors “fully mediates” the positive linkage between weather condition and stock market movement © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Example of mediating effect Does Superstitious Behaviors make people success? Superstitious Behaviors No direct relationship Success © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Example of mediating effect Superstitious Behaviors do not directly make people be successful. However, Superstitious Behaviors can help people increase confidence, which in turn, increases their chance of success. Po sit ive an ds tro ng Confidence Superstitious Behaviors No direct relationship an d str on g Success Increased confidence "fully mediates” the positive linkage between Superstitious Behaviors and Success © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Example of mediating effect Strong social connection directly increases chance of success. But why does strong social connection help people increase chance of success? Social connection Positive and strong Success © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Example of mediating effect Social connection allows people to gain access to important resources, thereby helping people increase the chance of success Social connection g Po s on str itiv ea d an nd ive sit Po str on g Access to important resources Positive and strong Success Access to important resources “partially mediates” the link between Social connection and Success © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Moderating effect vs Mediating effect Moderating effect Mediating effect Moderator can increase or reduce the effect of X on Y. Mediator links X and Y The degree to which X affects Y depends on a moderator X may not affect Y directly, but X affect Y through a mediator. © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Linear VS Nonlinear relationship • A linear relationship exists when two quantities are proportional to each other. • If we increase one of the quantities, the other quantity either increases or decreases at a constant rate. Positive (+) relationship Negative (-) relationship © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12– 51
Linear VS Nonlinear relationship • A nonlinear relationship is a type of relationship between two entities in which change in one entity does not correspond with constant change in the other entity. U-curve Inverse u-curve © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12– 52
Linear VS Nonlinear relationship • U-curve relationship • A small increase in X makes Y reduce (a relationship is negative) • When X increases above a certain point, it leads to the increase in Y (a relationship turn to be positive) Y 4 Example: The first few days of exercise makes our body weaker After that, our body become stronger 3 2 1 1 2 3 4 5 6 7 8 X © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 12– 53
Linear VS Nonlinear relationship • Inverse U-curve relationship • A small increase in X makes Y increase (a relationship is positive) • When X increases above a certain point, it leads to the decrease in Y (a relationship turn to be negative) Y Example: A few cups of coffee make us gain more energy More than that, our body begins to lose energy. 4 3 2 1 1 2 3 4 5 6 7 8 © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. X 12– 54
- Slides: 54