# MKT 500 Lecture Notes - Lecture 8: Frequency Distribution, Confidence Interval, Statistical Hypothesis Testing

51 views2 pages
15 Aug 2016
School
Department
Course
CMKT 500 - Lecture 8, 9 and 10 – Sampling, Sample Size, Fieldworker and data
quality issues, Analysis and Differences Testing
Sampling: Due to the unfeasibility of talking to the survey’s entire population, a small
portion of the target population is drawn from and treated as a representation of the whole.
Data Collection Errors:
Derive from fieldworkers, researchers, or the sample
Intentional field worker errors: intentionally incorporate some bias or error into data
collection (i.e. influence the respondent).
Unintentional fieldworker error: error in the data collection process is accidentally
created (i.e. mistake in the delivery of the survey)
Intentional respondent error: knowingly introduce errors into responses (i.e. biases)
Unintentional respondent error: error in the data collection process is accidentally
entered (i.e. wrongly interpreted question)
Online data collection: respondents can misrepresent themselves, submit multiple
versions of the same survey or give false responses
Nonresponse errors: survey that are not completed (refusal to participate, breakoffs
during the survey, not completing the survey)
Because: survey too long, questions become too intrusive or the respondent
might object to the questions
Reduce: pre-test the survey
Descriptive analysis: different forms allow research to present info logically and so it is
easily understood. Find some relationship in the data, identify meaningful difference, and
use simple charts (e.g. bar graphs, pie charts, line graphs)
Frequency distribution: shows the difference in the responses from one end to the other.
Central tendency: demonstrate the typical response to a question with mean/mode/median
Confidence interval: typically identified with a percentage. Researcher will try to identify the
parameters in which most of the population will fall then try to determine where x% of the
populations answers will fall (in a certain range).
Example. Test: determine average pizza delivery time. Track 500 orders. Create a
curve with the delivery time data. What time range do 95% of the times fall in? 25-30 mins,
therefore we could say any pizza delivery will take 25-30 with 95% confidence.
Hypothesis testing: used to accept or reject the hypothesis. Requires a sample population.
Hypothesis: statement of what the researcher thinks the population is doing/thinking,
but is unproven.
Null hypothesis: declares no relationship exists between the two ideas and groups or
between the population and sample. Determines whether or not a relationship exists.
Differences testing: compares answers/subgroups within a population.
find more resources at oneclass.com
find more resources at oneclass.com
Unlock document

This preview shows half of the first page of the document.
Unlock all 2 pages and 3 million more documents.

# Get access

\$10 USD/m
Billed \$120 USD annually
Homework Help
Class Notes
Textbook Notes