1--> X --> 2    

2 - 1 = Effect of X  + effect of 1

[1--->X]---2

2 - 1 = Effect of X + Effect of 1 + Effect of 1*X

X--> 1

Group watches ad, then gets surveyed. No pretest.

Lacks internal validity.

Simple, but you do not know what 1 means b/c there is nothing to base the survey off of.

Must establish a norm (prior examples, manager expectations, etc.)

1--->X--->2

Group is surveyed about attitude towards a brand, watches an ad, then takes post survey about their new attitude towards the brand.

History/Maturation effects

Know the level of effect of X = 2-1

knowledge at individual level (postive, negative and neutral opinions and what is driving them)

2-1 = Experimental effect + Extraneous Effect + MTE + ITE

Lack of randomization means poor external validity

A true experimental design!

Two Groups, Experimental Group & Control Group

EG: (R)-->1-->X-->2

CG: (R)-->3------->4

Size of effect:

EG: 2 - 1 = EE + ExE + MTE + ITE

CG: 4 - 3 = 0 + ExE + MTE + 0___

(2 - 1) - (4 - 3) = EE + ITE 

1) Response time

2) Cost per respondant

3) Quality of data

4) Flexibility in survey designing

1. Telephone Interviews
2. Electronic
3. MailInterviews
4. Personal Interviews

conducted in controlled setting, create a situation with desired conditions, manipulate some variables while controlling others, examine dependent variable

Advantages: control, internal validity

Disadvantages: external validity, not transferable to actual marketplace

Tests conducted outside the lab, reserach study in natural setting, manipulate some variables, try to find as controlled conditions as possible

Advantages: realism

Disadvantages: internal validity, no control over external factors

Researcher has control of variables and manipulates them:

-Treatment

-Subjects

-Dependent Variable

-Plan for extraneous causual factors

1. One-Shot Case Study
2. One Group Pretest Posttest
3. Static Group

EG: X ---> 1 [Exposure-->Observation]

CG: -----> 1 [Observation]

Preexperimental design with no pretests or randomization

Pretest-Posttest Control Group Design

Posttest-Only Control Group Design

Random assignment to EG and CG:

EG: (R) [Observation-->Exposure-->Observation]

CG: (R) [Observation----------------Observation]

Random assignment to test group, no pretest

EG: (R) [Exposure---Observatoin]

CG: (R) [Observation]

A series of observations, Exposure, another Series of observations.

Helps establish baseline for comparison after exposure occurs, but lacks internal validity because of History.

List advantages and disadvantages for the following four types of surveys methods:

1) Telephone Interview

2) Personal Interview

3) Mail Interview

4) Electronic Interview

1) Telephone Interview: 

• Personalized, large scale, quality control, honest responses, flexibility
• Lack of visual, time constraint, limited open-ended info, lower quality info

2) Personal Interview:

• Good response rate, better quality information, probing by interview, observe attitude
• Expensive, time consuming, may produce non-representative sample

3) Mail Interview: 

• Easy, cost efficient, honest responses
• Low response rate, no probing from interviewer, bad for illiterate audiences
  

4) Electronic Interview: 

• Low cost, automated, fast, convenient for respondents, design flexibility, honest responses
• Limited sample population, no probing from interviewer