text updates

Author: florian-huber

notebooks/11_correlation_analysis.ipynb

@@ -374,9 +374,9 @@
    "id": "02555327",
    "metadata": {},
    "source": [
-    "**More specific example**\n",
+    "**More specific (generated) example**\n",
     "\n",
-    "In the following we still look at generated data, but data which is somewhat realistic. We want to see if body measures show any interesting correlations, say between shoe size and height of people."
+    "In the following, we continue to work with generated data, but now data that is at least somewhat realistic. We want to see if body measures show any interesting correlations, and we start with (fake) data containing shoe size and height of people."
    ]
   },
   {
@@ -489,7 +489,7 @@
    "id": "fe842f17",
    "metadata": {},
    "source": [
-    "Let's first do some standard inspection of the data."
+    "As it was introduced in the previous sections, we would often do a quick first inspection of the data using simple statistical measures. In Pandas this is very easy to do:"
    ]
   },
   {
@@ -605,7 +605,9 @@
    "source": [
     "**Basic cleaning or processing**\n",
     "\n",
-    "In the table above, some things seem weird. The minimum height is 1.63 and the maximum height is 364.0. Probably some fantasy figures, one apparently also with shoe size 88. This are typical issues we can quickly discover in a first inspection and then decide what to do about it. Here, we will simply decide to only take data within more or less realistic boundaries. "
+    "In the table above, some things seem weird!\n",
+    "\n",
+    "The minimum height is 1.63, and the maximum height is 364.0. Something is not correct here. Maybe someone added some fantasy figures to the dataset, one of them apparently with a shoe size of 88. These are typical issues we can quickly discover in a first inspection. We then have to decide what to do about it. Here, we simply decide to  keep data only within more or less realistic boundaries. "
    ]
   },
   {
@@ -624,7 +626,7 @@
    "id": "003e6b3c",
    "metadata": {},
    "source": [
-    "We can then move on to the correlations:"
+    "We can then move on to computing the correlations:"
    ]
   },
   {