Program features
The GUI displayed in the image below consists of a single display made up of three panels. In panel 1 and 2, the parameters for the specific study are set. In panel 3, the simulations are prompted and its results are displayed. In the following, the panels are explained.
-
- Number of attributes per person
Program features
Interpretation of results
The following image shows an enlargement of the lower part of panel 3.
-
The graphic contains three facets, with one facet for each requested proportion value. In each facet, the x-axis shows the simulated sample sizes N and the y-axis the probability of getting the specific result. For each value M, i.e. the minimal number of constructs contained in a category, a separate line is drawn, as indicated in the legend on the right. To answer the question
“What is the probability to get a result where at least C=95% percent of the categories contain a minimum of M=5 attributes when using the sample size 50?”
@@ -148,7 +148,7 @@Interpretation of results
“What is minimum sample size required, to get a result where C=100% percent of the categories contain a minimum of M=1 attributes (i.e. are recovered) with a probability of almost 100%?”
It is not necessary to rerun the simulation but simply change the drawing parameters. To request a suitable graphic to evaluate this question we can set the minimum count to \(M=1\) and use a proportion of \(C=`.95, 1`\). The result is displayed in the figure below. It can be seen that the probability to sample a construct from every category when using \(N=40\) is close to \(1\). With \(N=20\) we can still be confident to recover all categories with a probability of \(.75\). This is somewhat different from the rule of thumb of \(N \in [15; 25]\) to recover all categories given in the literature (Kerkhof et al., 2009; Napier et al., 2009; Tan & Hunter, 2002) and shows that study specific simulations may be useful.
-
User Feedback
diff --git a/inst/doc/index.html b/inst/doc/index.html index de17205..90da5c3 100644 --- a/inst/doc/index.html +++ b/inst/doc/index.html @@ -108,7 +108,7 @@Install and load
Program features
The GUI displayed in the image below consists of a single display made up of three panels. In panel 1 and 2, the parameters for the specific study are set. In panel 3, the simulations are prompted and its results are displayed. In the following, the panels are explained.
-
- Number of attributes per person
Program features
Interpretation of results
The following image shows an enlargement of the lower part of panel 3.
-
The graphic contains three facets, with one facet for each requested proportion value. In each facet, the x-axis shows the simulated sample sizes N and the y-axis the probability of getting the specific result. For each value M, i.e. the minimal number of constructs contained in a category, a separate line is drawn, as indicated in the legend on the right. To answer the question
“What is the probability to get a result where at least C=95% percent of the categories contain a minimum of M=5 attributes when using the sample size 50?”
@@ -148,7 +148,7 @@Interpretation of results
“What is minimum sample size required, to get a result where C=100% percent of the categories contain a minimum of M=1 attributes (i.e. are recovered) with a probability of almost 100%?”
It is not necessary to rerun the simulation but simply change the drawing parameters. To request a suitable graphic to evaluate this question we can set the minimum count to \(M=1\) and use a proportion of \(C=`.95, 1`\). The result is displayed in the figure below. It can be seen that the probability to sample a construct from every category when using \(N=40\) is close to \(1\). With \(N=20\) we can still be confident to recover all categories with a probability of \(.75\). This is somewhat different from the rule of thumb of \(N \in [15; 25]\) to recover all categories given in the literature (Kerkhof et al., 2009; Napier et al., 2009; Tan & Hunter, 2002) and shows that study specific simulations may be useful.
-
User Feedback
diff --git a/inst/shiny/text/index.html b/inst/shiny/text/index.html index e9f5f0d..a536783 100644 --- a/inst/shiny/text/index.html +++ b/inst/shiny/text/index.html @@ -63,7 +63,7 @@Install and load
Program features
The GUI displayed in the image below consists of a single display made up of three panels. In panel 1 and 2, the parameters for the specific study are set. In panel 3, the simulations are prompted and its results are displayed. In the following, the panels are explained.
-
- Number of attributes per person
Program features
Interpretation of results
The following image shows an enlargement of the lower part of panel 3.
-
The graphic contains three facets, with one facet for each requested proportion value. In each facet, the x-axis shows the simulated sample sizes N and the y-axis the probability of getting the specific result. For each value M, i.e. the minimal number of constructs contained in a category, a separate line is drawn, as indicated in the legend on the right. To answer the question
“What is the probability to get a result where at least C=95% percent of the categories contain a minimum of M=5 attributes when using the sample size 50?”
@@ -103,7 +103,7 @@Interpretation of results
“What is minimum sample size required, to get a result where C=100% percent of the categories contain a minimum of M=1 attributes (i.e. are recovered) with a probability of almost 100%?”
It is not necessary to rerun the simulation but simply change the drawing parameters. To request a suitable graphic to evaluate this question we can set the minimum count to \(M=1\) and use a proportion of \(C=`.95, 1`\). The result is displayed in the figure below. It can be seen that the probability to sample a construct from every category when using \(N=40\) is close to \(1\). With \(N=20\) we can still be confident to recover all categories with a probability of \(.75\). This is somewhat different from the rule of thumb of \(N \in [15; 25]\) to recover all categories given in the literature (Kerkhof et al., 2009; Napier et al., 2009; Tan & Hunter, 2002) and shows that study specific simulations may be useful.
-