Adv 344k Exam 3
report ANOVA
"A one-way between subjects ANOVA was conducted to compare the effect of (IV)______________ on (DV)_______________ in _________________, __________________, and __________________ conditions." "There was a significant (not a significant) effect of IV ____________ on DV ______________ at the p<.05 level for the three conditions [F(___, ___) = ___, p = ____].
t test
"An independent-samples t-test was conducted to compare (your DV measure) _________ in (IV level / condition 1) ________and (IV level / condition 2)________ conditions. "There was a significant (not a significant) difference in the scores for IV level 1 (M=___, SD=___) and IV level 2 (M=___, SD=___) conditions; t(__)=____, p = ____"
post hoc
"Post hoc comparisons using the Tukey HSD test indicated that the mean score for the sugar condition (M = 4.20, SD = 1.30) was significantly different than the no sugar condition (M = 2.20, SD = 0.84). However, the a little sugar condition (M = 3.60, SD = 0.89) did not significantly differ from the sugar and no sugar conditions."
Answers to questions in Assignment 4
"were all of the relationships significant?" general questions of the assignment
unintentional error
- 5 types -respondent misunderstanding -guessing -attention loss -distractions -respondent fatigue
Marketing research reports, objectives
-To effectively communicate the findings of the marketing research project -To provide interpretations of those findings in the form of sound and logical recommendations -To establish the credibility of the research project -To serve as a future reference document for strategic or tactical decisions
guidelines for oral and visual presenting
-Visual component should not detract from the information being communicated -Be friendly, honest, warm, and open in your oral communication -Delivery should be knowledgeable and confident -Have a well-organized and inspiring dialogue prepared -Be an effective active listener, let them interrupt
questionaire design has 6 key functions
-What should be asked? -How should questions be phrased? -In what sequence should the questions be arranged? -What questionnaire layout works best? -How should the questionnaire be pre-tested? -Does the questionnaire need to be revised?
Elements of the report
-front matter -intro -research obj -method -methodology -results -limitiations -conclusions + recommendations
examples of conceptualization
-profanity -definitons
Beyond organization, there are guidelines that will ensure that the data presentation tables strongly imply that they are credible and should be taken very seriously. Which of the following is NOT one of those guidelines? a. With scales, include a table footnote that describes the scale. b. Always use two-three colors in each table c. Use one decimal place unless convention demands otherwise. d. Only report findings that are meaningful or useful
. Always use two-three colors in each table
levels of measurement
1 nominal: category, name, label (male = 0, female = 1) can't average, use counts. 2 ordinal: ordered from least to most, but distance between isn't precise (fav restaurant 1-5) 3 interval: distance between attribute does have meaning: no zero point (strong disagree, somewhat disagree, etc...) 4 ratio: absolute zero point that is meaningful (how angry from 0-100)
Ways to administer surveys
1. Conducted in person, one-on-one (home, mall intercept) 2. Conducted by telephone (often uses a CATI system - computer assisted telephone interviewing) 3. Group-administered in person (e.g., in waiting room) 4. Mailed via U.S. Mail, or dropped off in person 5. Panel participation (Mail or online) 6. Conducted electronically via email or the Internet.
Response Rate and How to Calculate
1. Response rate refers to percentage of valid sample who provide a valid, completed interview -Calculated by dividing number of respondents completing survey by total number of respondents in the VALID sample
chi square report
A chi-square test of independence was performed to examine the relation between religion and college interest. The relation between these variables was significant, X2 (2, N = 170) = 14.14, p <.01. Catholic teens were less likely to show an interest in attending college than were Protestant teens.
Chi-square test
A significance test used to determine if a linear relationship exists between two variables measured on interval or ratio scales. Chi-square stat see if there are significant differences in the frequency distribution
outlier
A value much greater or much less than the others in a data set
Variability (variance)
Degree of difference or deviation from mean.
validity
Did we we measure what we intended intended to measure?
Frequency distribution, tables, graphs
Frequency distribution table shows how many people (number & percentage) answered in each response category., can also be displayed as a graph
disadvantages of surveys
Generally limited to scale or check listed questions. No control over response rate. Provides information more than understanding. Difficult to explore in depth issues.
IRB
Institutional Review Board, review research in advance to ensure ethical considerations are met
measurement process (starts with concept
Once you have decided which concept is important to study, it is time to figure out how measure its presence or absence measure when it does or doesnt occur = called operationalization
likert
Scale that primarily measures attitudes by asking respondents their degree of agreement or disagreement to a number of statements
margin of sampling error
The degree of inaccuracy in any poll, arising from the fact that surveys involve a sample of respondents from a population, rather than every member. accuracy increases with greater and greater sample sizes but there is only a minute gain in accuracy when the sizes are more than 1000 respondents
What is the interval scale that has mainly degrees of positive positions?
The nonsymmetric interval scale
Central Limit Theorem
The theory that, as sample size increases, the distribution of sample means of size n, randomly selected, approaches a normal distribution. - if a replication is repeated lots of times and a bar chart of all percents of all results, the central limit theorem holds that it would look like a normally distributed curve
Excel, use for descriptives
To calculate descriptive stats, need to use this function (fx).
questions to ask in conceptualization
What is the concept? how will it surface and vary within the sample (the variables)? how will we measure it with indicators? -google search -google scholar search
Census education example
What is the highest grade of school you have has completed, or the highest degree you have received?
Indices, sources for, Handbook
When several questions are used to measure one concept, the responses can be summed or averaged. This is called an index. handbook has many indices that can be searched through and utilized (think in class assignment)
nonsampling error
a bias that occurs in a research study regardless of whether a sample or census is used -all sources of error other than sample selection methd and sample size. (includes problem specific mistakes, question bias, data recording errors, incorrect analysis)
semantic differential scale
a five-point scale in which the opposite ends have one- or two-word adjectives that have opposite meanings
convenience sample
a form of nonprobability sample using respondents who are convenient or readily accessible to the researcher—for example, employees, friends, or relatives
sampling frame
a list of individuals from whom the sample is drawn
purposive sample
a nonrandom sample that is chosen for some characteristic that it possesses requires educated guess at who should represent the population
behavioral intention scale
a special type of rating scale designed to capture the likelihood that people will demonstrate some type of predictable behavior intent toward purchasing an object or service in a future time frame
sample
a subset of the population
stratified sampling
a variation of random sampling; the population is divided into subgroups and weighted based on demographic characteristics of the national population
Marketing research reports are ________, and you must consider both in all phases of the research process, including planning the report. a) tailored to specific audiences and purposes b) written for both paper and electronic presentations c) written for mid- and upper-level management d) primarily focused on communicating statistics and meaning of charts
a)) tailored to specific audiences and purposes
what to determine before sampling
accuracy of sampling frame and difference between those who chose to respond vs non-respondents (determine population, what you hope to determine, what needs to be represented, etc)
census
an accounting of the complete population- no sample taken, everyone is included
sampling error
an error that occurs when a sample somehow does not represent the target population -error in a survey that occurs because a sample is used.either due to method of sample selection, or sample size.
concept
an idea or term that summarizes similar observations or experiences
When is one item ok?
asking what your gender is, if you support presidential candidate, etc obviously for complex ideas like attachment you can't just have a single item
association vs. cause and effect
association means that two variables are related, have a correlation. does not mean that one variable CAUSES the other.
There are statistical analyses beyond simple descriptive measures, statistical inference, and differences tests including ___, which determine whether a stable relationship exists between two variables. a) Analysis of variance analyses b) Associative analyses c) Predictive analyses d) Regression analyses
associative analyses
Confidentiality
assurance that respondents answers are private
The importance of the marketing research report cannot be understated; even the best research will not drive the appropriate action unless: a) The audience understands the results and their meaning. b) The marketing researcher is involved in presenting the results. c) Top management reads the report in its entirety. d) The statistics are properly presented and explained.
audience understand result and meaning
The first item that every market researcher should consider when preparing his or her presentation is: a) Plan a coherent ending that summarizes findings b) Identify and analyze the audience c) Determine key points the audience needs to hear d) Make sure visuals portray key points
b) Identify and analyze the audience
________ are useful to compare items over time or show correlations among items. a) Pie charts b) Bar charts c) Flow diagrams d) tables
bar charts
There are several guidelines in the presentation of hypothesis tests. Which of the following does not represent one of those guidelines? a. Perform appropriate hypothesis test computations b. Determine if the hypothesis is supported or not supported by comparing the computed level of confidence c. Prepare hypothesis test results in both print and slide formats d. state the hypothesis
c. Prepare hypothesis test results in both print and slide formats
A researcher accumulates the gender of each respondent. Which of the following descriptive statistics would he or she likely use? a. range b. standard deviation c. frequency and/or percentage distribution d. cumulative percentage distribution
c. frequency and/or percentage distribution
exhaustive
captures all levels of the variable
incentives
cash payments, gift cards, something of value promised to respondents for participating
Central tendency and variation
central tendency measured by mean, median, mode variation measured by variance(average of squared differences from the mean), range (highest-lowest value), and standard deviation (how spread out from the mean (VARIANCE AND SD DIFF
The degree to which the observed frequencies depart from the expected frequencies is expressed in a single number called the: a) Z-test statistic b) Analysis of variable statistic c) Chi-square test statistic d) T-test statistic
chi square test
closed ended
closed ended questions must be exhaustive to minimize people who select "other"
random number generator
computer programs to give random numbers so they can draw a huge random sample from a gigantic population in minutes to be sure that every pop member has same chance of being selected
relability
consistent scores over time and contexts. effected less by random error.
Cost/Timing/Needs/Length
cost is affected by design manpower needed materials needed length analysis can be expensive for random sample telephone/ online takes least tinme mail and in person is slowest personal contact is necessary for long length surveys online/tele caters to short surveys
If you found a significant correlation of .83 between years of education and hours spent attending San Francisco Giants' baseball games, it would mean that: a) People with less education spend less time attending Giants' baseball games b) People with more education spend less time attending Giants' baseball games c) People with less education spend more time attending Giants' baseball games d) People with more education spend more time attending Giants' baseball games
d) People with more education spend more time attending Giants' baseball games
Which of the following is best suited for allowing users to quickly and easily see information that is presented in a simplified manner? a) Online research reports b) Infographics c) Social media d) Dashboards
dashboards
deductive approach
define research objectives very early in the process rather than inductive inductive results are assumed through data, deductive seeks to prove ideas through data collection/ interpretation
Certain measures such as the mean, mode, and standard deviation, and range, are forms of ____ used by marketing researchers to describe the sample dataset in such a way as to portray the "typical" respondent and to reveal the general pattern of responses. a. association analysis b. descriptive analysis c. inference analysis d. difference analysis
descriptive stats
p value, sig
determines if means are significant;y different p<.05 it is significantly different statistically
____ means the researcher knows if the relationship between variables is positive or negative. a) direction b) pattern c) general d) significance
direction
When a survey displays or asks questions that are appropriate based on the respondent's prior answers, this is known as:
display logic
cluster smapling
divide pop into clusters that each represent population and use as sample
nonresponse error
due to respondents not answering questions
population
entire group under study
simple random sample
every member of the population has a known and equal chance of selection
explanatory/ exploratory
experimental design where half of the sample gets one stimuli and half gets another
There are several sections of a research report. Which of the following is NOT one of the basic sections? a) front matter b) end matter c) graphs and charts d) body
graphs and charts
Regardless of its absolute value, a correlation that is not statistically significant: a) Has significant meaning b) Has some perceptible meaning c) Requires additional examination d) Has no meaning at all
has no meaninf
The comparison of the sample statistic with what is believed to be the population value prior to undertaking the study is called: a. parameter testing b. population testing c. inference testing d. hypothesis testing
hypothesis testing
codebook
identifies the questions, the variable name or label associated with each question, and the code of numbers associated with each possible response to each question know how to make sense of the results
representative sample
if we want to generalize about the population, we must have a representative sample: must mirror the population in all respects that are potentially relevant to the study. (if it is unrepresentative, some elements are over represented or underrepresented)
Depending on its type, a relationship can usually be characterized by all of the following EXCEPT: a) its absence b) its strength of association c) its presence d) its direction
its absence
Index example of "product involvement" and "attatchment"
know generally what they measure) measure attachment to vehicle in 3 stages of ownership involvement with a product through semantic differential scale
advantages of surveys
large sample size generalizable precise estimates to identify small differences easy to admin and record answers advanced statistical analysis concepts/ relationships not directly measurable/ can be studied
Cronbach's alpha, interpreting
less than .6 unacceptable .6-.65 undesirable .66-.7 minimally acceptable .71-.8 respectable .81-.9+ very good when considering inter-term reliability we calculate correlation between 0 and 1.
nonprobability
members of the population do NOT have a known chance (probability) of being selected into the sample
probability
members of the population have a known chance (probability) of being selected into the sample
With a nominal scale, the ___ is the appropriate measure of central tendency, and variation must be assessed by looking at the distribution of responses across the various response categories. a. variable b. mean c. median d. mode
mode
improving response rate
motivating invitation incentives functional links reassure confidentiality keep it short take care with to from subject fields
Distribution, normal, skewed
normal: shows frequency of means. 68% within 1 s.d. of mean, 95% of data within 2 s.d., 99.7% within 3 s.d. of mean. skewed: has a tail right or left, data is not concentrated right in the center.
Hypotheses, hypothesis testing, null
null hypothesis is that there is no difference between gender groups. The alternative hypothesis is that there is. hypothesis test is a statistical procedure used to "accept" or reject the hypothesis based on sample evidence
Video footage that is gathered through ________ can be used to illustrate key takeaways from research studies. a) Observation or ethnographic methods b) Socio-cultural methods c) proprietary film databases d) live news broadcasts
observation or ethnographic methods
open ended
open ended questions are hard to code for since people may respond in completely unique ways.
Post-hoc tests
options that are available to determine where the pair(s) of statistically significant differences between the means exist(s) after ANOVA
Bivariate Pearson correlations
pearson correlations between two variables (report in terms of strength, direction, significance) "involvement with a product is positively related with intention to purchase that product"
elements
people in the samples
Which of the following is particularly useful for illustrating relative size or static comparisons? a) pie charts b) bar charts c) tables d) flow diagrams
pie charts
A ________ involves conducting a dry run of the survey on a small, representative set of respondents to reveal questionnaire errors before the survey is launched.
pretest
Percentages, calculating
proportion or share in relation to a whole
The ___ specifies the difference between the endpoints in a set of values arranged in order. a. mode b. mean c. standard deviation d. range
range
The type of measure in which a respondent responds in the same or very similar manner to an identical or near-identical question is called:
reliable mesaure
refferal sample
require respondents to provide the names of prospective respondents -aka snowball sampling
mutually exclusive
response options dont overlap
Errors in Surveys
sampling errors and non-sampling errors
A ____ allows a researcher to visually inspect the plotted points and possibly to spot a systematic nonlinear relationship. a) multiple line graph b) line graph c) Stacked bar chart d) scatter plot (Scatter diagram)
scatter plot diagram
systematic sampling
select some starting point and then select every kth element in the population
Handbook of Marketing Scales
semantic is most common
Sig., strength, direction
significance= p value strength= how spread out or clustered is data (between 0 and 1) direction= positive or negative correlation value?
Crosstab, interpreting
simple method for looking at statistically significant differences, used for relationships among nominal variables or to compare groups (distribution of one variable for each category for another variable)
quota sampling
specified percentages of the total sample for various types of individuals to be interviewed
descriptive statistics
statistical procedures used to describe characteristics and responses of groups of subjects statistics that summarize the data collected in a study
Values that are computed from information provided by a sample are referred to as the sample's: a. statistics b. variabilities c. parameters d. errors
stats
what is a survey
survey is a tool to collect data from a sample of individuals, that can collect quantatitive data for statistical analysis and uses closed-ended questions to record self-report responses. popular for their versatility, generalizable , and good for measuring attitudes
descriptive
surveys tend to be descriptive: focused on describing behaviors or attitudes and looking for relationships between variables
Split AB Testing
test different website to see what is most effective
independent samples t test
test that tries to determine if there is statistically significant difference between two groups' means
sampling units
the basic level of investigation (individuals, classes, etc.)
sample frame error
the degree to which the sample frame fails to account for the population
Generalizability
the extent to which we can claim our findings inform us about a group larger than the one we studied the extent to which your sample can be said to characterize your population.
data coding
the identification of code values that are associated with the possible responses for each question on the questionnaire
sample size to sampling error
the larger a random sample is, the more accurate it is and it has a smaller margin or sample error
confidence level
the percentage of area under the normal curve described by our calculated confidence intervals
operationalization
the process of assigning a precise method for measuring a term being examined for use in a particular study -what is the concept -how will it surface and vary within the sample -how will we measure it with indicators
conceptualization
the process of isolating the concepts that make up a phenomenon
confidence interval
the range of values within which a population parameter is estimated to lie Confidence interval, level a range whose endpoints define a certain percentage of the responses to a question. based on a normal bell-shaped curve .
nonsampling error
the result of poor sample design, sloppy data collection, faulty measuring instruments, bias in questionnaires, and so on. -all types of nonresponse error, data-gathering error, data-handling error, data analysis error, and interpretation error
target population
the total group to be studied or described and from whom samples may be drawn what do you need to know, how are you thinking about your target populations and what factors need to be considered
bimodal
two peaks
random digit dialing
used in telephone surveys to avoid the problems of unlisted cellphone numbers
Analysis of variance (ANOVA)
used to compare multiple groups. differences between the group means to ascertain whether sampling errors or true population differences explain their failure to be equal. Helps us find out if the IV had a significant effect on the DV. doesn't tell us which two groups are significantly different
Correlation coefficients that fall between the absolute values of 1.00 and .81 are generally considered to be: a) Very weak b) Somewhat strong c) Very strong d) Somewhat
very storng
intentional error
willfully misrepresenting themself in survey -falsehood is when they lie -nonresponse skips question
question bias
xthe ability of a question's wording to influence respondents' answers. This needs to be minimized