Download MS-PowerPoint - People Server at UNCW

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Probability wikipedia , lookup

History of statistics wikipedia , lookup

Foundations of statistics wikipedia , lookup

Statistics wikipedia , lookup

Transcript
THE IMPACT OF [INDEPENDENT VARIABLE]
ON [DEPENDENT VARIABLE]
CONTROLLING FOR [CONTROL VARIABLE]
[Your Name]
PLS 401, Senior Seminar
Department of Public & International Affairs
UNC Wilmington
5/25/2017
1
Univariate Hypothesis
• Theory:
–X
–X
• H1: predict the distribution of values across the categories of
your dependent variable. If relevant, predict whether you
expect to find a conflict or consensus distribution.
5/25/2017
2
Table 1
[insert the SETUPS frequency table
for your dependent variable]
5/25/2017
3
Univariate Findings
• H1 ([restate hypothesis]) is [supported/ not supported /
contradicted] by the sample data in Table 1 because:
1. The pattern predicted by H1 [is/is not observed in/is
contradicted by] the sample data.
2. The pattern observed in the sample [is/is not] statistically
significant. The random-sampling error margin for this size
sample is [± x %].
5/25/2017
4
Bivariate Hypothesis
• Theory:
–X
–X
• H2: [one category of the independent variable] is more likely
than [another category of the independent variable] to
[exhibit a particular value of the dependent variable]. [for
example: males are more likely than females to support the
death penalty – where gender is the independent variable and
attitude toward the death penalty is the dependent variable]
5/25/2017
5
Table 2:
[insert the bivariate SETUPS table and
include the tau-b & chi-squared probability statistics]
5/25/2017
6
Bivariate Findings
• H2 ([restate the bivariate hypothesis]) is [supported/ not
supported/is contradicted] by the sample data in Table 2
because:
1. The pattern predicted by H2 [is/is not] observed in the
sample data. The tau-b is [x.xx] which indicates that the
relationship is [weak/moderate/strong].
2. This sample finding [is/is not] statistically significant. The
chi-squared probability of random-sampling error [is/is not]
less than 0.05 (it is [x.xx]).
5/25/2017
7
Multivariate Hypothesis
• Theory:
– X
– X
• H3: controlling for [the control variable] [does / does not] change the
impact of [the independent variable] on [the dependent variable] across
the partial tables.
– In the [first partial-table subgroup], the bivariate relationship will be
[weaker / the same / stronger] than in the total population.
– In the [second partial-table subgroup], the bivariate relationship will
be [weaker / the same / stronger] than in the total population.
– Add a prediction for the 3rd partial-table subgroup, if necessary.
5/25/2017
8
Table 3a
[insert the first SETUPS partial table and
include the tau-b & chi-squared probability statistics]
5/25/2017
9
Table 3b
[insert the second SETUPS partial table and
include the tau-b & chi-squared probability statistics]
5/25/2017
10
Table 3c
[if necessary, otherwise delete this slide]
[if necessary, insert the third SETUPS partial table and
include the tau-b & chi-squared probability statistics]
5/25/2017
11
Multivariate Findings
• H3 ([restate the multivariate hypothesis)] is [supported / not
supported / contradicted] by the sample data.
1. The strength of the bivariate relationship [did / did not]
change as predicted in the partial-table subgroups. [Report
and interpret the tau-b statistics]
2. The statistical significance of the bivariate relationship [did /
did not] change in the partial-table subgroups. [Report and
interpret the chi-squared probability statistics]
5/25/2017
12
Substantive Implications
• Suggest several implications of these findings for political
decision makers and government officials.
• X
• X
5/25/2017
13
Methodological Implications
• Suggest several implications of these findings for other
researchers interested in this topic.
• X
• X
5/25/2017
14
References
• x
• Shively, W. Phillips. 2008. Power & Choice: An Introduction to
Political Science. 11e. Boston: McGraw-Hill.
• x
5/25/2017
15