cite

alexanderthclark · Apr 17, 2024 · 581096a · 581096a
1 parent cbe32dd
commit 581096a
Show file tree

Hide file tree

Showing 8 changed files with 19 additions and 19 deletions.
diff --git a/.DS_Store b/.DS_Store
diff --git a/book/_build/.doctrees/chi2.doctree b/book/_build/.doctrees/chi2.doctree
diff --git a/book/_build/.doctrees/environment.pickle b/book/_build/.doctrees/environment.pickle
diff --git a/book/_build/html/_sources/chi2.ipynb b/book/_build/html/_sources/chi2.ipynb
@@ -680,7 +680,7 @@
     "1. Counted (qualitative) Data \n",
     "2. Independence (e.g. individuals being counted are sampled independently)\n",
     "3. Randomization (to generalize results)\n",
-    "4. Expected Cell Frequency (rule of thumb: expectation of at least 5 for each cell)\n",
+    "4. Expected Cell Frequency (rule of thumb: expectation of at least 5 for each cell {cite}`destats`)\n",
     "\n",
     "### Constructing Contingency Tables in Google Sheets\n",
     "\n",
@@ -806,7 +806,7 @@
     "\n",
     "Is the test statistic large enough to lead us to reject the null? There are 2 degrees of freedom, so the critical value would be 5.99 according to {numref}`chi2Table`. We can reject the null because our statistic is greater. We have evidence that genre and song theme are related. It seems like Swift's genre shift also included thematic shifts.\n",
     "\n",
-    "But can we trust the analysis? Our calculations are correct but we should think about the assumptions. Our data is small. In particular, the expected cell count for Country and Self-Discovery is just 2.56. A rule of thumb is that each expected cell count should be at least five. Our test fails this rule of thumb, so we shouldn't take the analysis that seriously. Further, even if we leave aside the sample size, we must include the usual caveats about causality. There's nothing in the data or the test that the genre drives different song themes. Swift's transition to pop also came with her maturation as an artist. Perhaps more mature artists progress to different song themes and that is related to the the genre shift only coincidentally. \n",
+    "Can we trust the analysis? Our calculations are correct, but we should think about the assumptions. Our data is small. In particular, the expected cell count for Country and Self-Discovery is just 2.56. A rule of thumb is that each expected cell count should be at least five. Our test fails this rule of thumb, so we shouldn't take the analysis that seriously. Further, even if we leave aside the sample size, we must include the usual caveats about causality. There's nothing in the data or the test that the genre drives different song themes. Swift's transition to pop also came with her maturation as an artist. Perhaps artists progress to different song themes over time and that is related to the the genre shift only coincidentally. \n",
     "\n",
     "##### Python Code\n",
     "\n",
@@ -816,7 +816,7 @@
   {
    "cell_type": "code",
    "execution_count": 9,
-   "id": "7538d0cf",
+   "id": "7b04429d",
    "metadata": {},
    "outputs": [
     {

diff --git a/book/_build/html/chi2.html b/book/_build/html/chi2.html
diff --git a/book/_build/html/searchindex.js b/book/_build/html/searchindex.js
diff --git a/book/_build/jupyter_execute/chi2.ipynb b/book/_build/jupyter_execute/chi2.ipynb
diff --git a/book/chi2.ipynb b/book/chi2.ipynb
@@ -680,7 +680,7 @@
     "1. Counted (qualitative) Data \n",
     "2. Independence (e.g. individuals being counted are sampled independently)\n",
     "3. Randomization (to generalize results)\n",
-    "4. Expected Cell Frequency (rule of thumb: expectation of at least 5 for each cell)\n",
+    "4. Expected Cell Frequency (rule of thumb: expectation of at least 5 for each cell {cite}`destats`)\n",
     "\n",
     "### Constructing Contingency Tables in Google Sheets\n",
     "\n",
@@ -816,7 +816,7 @@
   {
    "cell_type": "code",
    "execution_count": 9,
-   "id": "3b2f81c7",
+   "id": "7b04429d",
    "metadata": {},
    "outputs": [
     {