{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": { "collapsed": true }, "outputs": [], "source": [ "import warnings\n", "warnings.simplefilter('ignore')\n", "\n", "import numpy as np\n", "import pandas as pd\n", "import seaborn as sns\n", "import matplotlib.pyplot as plt\n", "import graphviz\n", "from sklearn import tree\n", "from sklearn.tree import DecisionTreeClassifier\n", "from sklearn.preprocessing import LabelEncoder\n", "\n", "%matplotlib inline" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Load Dataset" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
employee_idcompany_iddeptsenioritysalaryjoin_datequit_date
013021.07customer_service2889000.02014-03-242015-10-30
1825355.07marketing20183000.02013-04-292014-04-04
2927315.04marketing14101000.02014-10-13NaT
3662910.07customer_service20115000.02012-05-142013-06-07
4256971.02data_science23276000.02011-10-172014-08-22
\n", "
" ], "text/plain": [ " employee_id company_id dept seniority salary join_date \\\n", "0 13021.0 7 customer_service 28 89000.0 2014-03-24 \n", "1 825355.0 7 marketing 20 183000.0 2013-04-29 \n", "2 927315.0 4 marketing 14 101000.0 2014-10-13 \n", "3 662910.0 7 customer_service 20 115000.0 2012-05-14 \n", "4 256971.0 2 data_science 23 276000.0 2011-10-17 \n", "\n", " quit_date \n", "0 2015-10-30 \n", "1 2014-04-04 \n", "2 NaT \n", "3 2013-06-07 \n", "4 2014-08-22 " ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data = pd.read_csv('./data/employee_retention_data.csv', parse_dates=['join_date', 'quit_date'])\n", "data.head()" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "RangeIndex: 24702 entries, 0 to 24701\n", "Data columns (total 7 columns):\n", "employee_id 24702 non-null float64\n", "company_id 24702 non-null int64\n", "dept 24702 non-null object\n", "seniority 24702 non-null int64\n", "salary 24702 non-null float64\n", "join_date 24702 non-null datetime64[ns]\n", "quit_date 13510 non-null datetime64[ns]\n", "dtypes: datetime64[ns](2), float64(2), int64(2), object(1)\n", "memory usage: 1.3+ MB\n" ] } ], "source": [ "data.info()" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
employee_idcompany_idsenioritysalary
count24702.00000024702.00000024702.00000024702.000000
mean501604.4035303.42696914.127803138183.345478
std288909.0261012.7000118.08952076058.184573
min36.0000001.0000001.00000017000.000000
25%250133.7500001.0000007.00000079000.000000
50%500793.0000002.00000014.000000123000.000000
75%753137.2500005.00000021.000000187000.000000
max999969.00000012.00000099.000000408000.000000
\n", "
" ], "text/plain": [ " employee_id company_id seniority salary\n", "count 24702.000000 24702.000000 24702.000000 24702.000000\n", "mean 501604.403530 3.426969 14.127803 138183.345478\n", "std 288909.026101 2.700011 8.089520 76058.184573\n", "min 36.000000 1.000000 1.000000 17000.000000\n", "25% 250133.750000 1.000000 7.000000 79000.000000\n", "50% 500793.000000 2.000000 14.000000 123000.000000\n", "75% 753137.250000 5.000000 21.000000 187000.000000\n", "max 999969.000000 12.000000 99.000000 408000.000000" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.describe()" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "employee_id 0\n", "company_id 0\n", "dept 0\n", "seniority 0\n", "salary 0\n", "join_date 0\n", "quit_date 11192\n", "dtype: int64" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Null information\n", "data.isnull().sum()" ] }, { "cell_type": "markdown", "metadata": { "collapsed": true }, "source": [ "# Create Table for `day`, `employee_headcount`, and `company_id`" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "collapsed": true }, "outputs": [], "source": [ "# Define useful information\n", "unique_date = pd.date_range(start='2011-01-24', end='2015-12-13', freq='D')\n", "unique_company = sorted(data['company_id'].unique())\n", "\n", "day = []\n", "company = []\n", "headcount = []\n", "\n", "# Loop through date and company id\n", "for date in unique_date:\n", " for idx in unique_company:\n", " total_join = len(data[(data['join_date'] <= date) & (data['company_id'] == idx)])\n", " total_quit = len(data[(data['quit_date'] <= date) & (data['company_id'] == idx)])\n", " day.append(date)\n", " company.append(idx)\n", " headcount.append(total_join - total_quit)\n", "\n", "# Create table for day, employee_headcount, company_id\n", "table = pd.DataFrame({'day': day, 'company_id': company, 'employee_headcount': headcount}, \n", " columns=['day', 'company_id', 'employee_headcount'])" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
daycompany_idemployee_headcount
02011-01-24125
12011-01-24217
22011-01-2439
32011-01-24412
42011-01-2455
\n", "
" ], "text/plain": [ " day company_id employee_headcount\n", "0 2011-01-24 1 25\n", "1 2011-01-24 2 17\n", "2 2011-01-24 3 9\n", "3 2011-01-24 4 12\n", "4 2011-01-24 5 5" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "table.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Employee Churn Analysis" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "collapsed": true }, "outputs": [], "source": [ "# Separate stay and quit data\n", "quit_data = data[~data['quit_date'].isnull()]\n", "stay_data = data[data['quit_date'].isnull()]" ] }, { "cell_type": "markdown", "metadata": { "collapsed": true }, "source": [ "### Feature Engineering" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAA40AAAGICAYAAADlBgWjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Total working days\n", "work_days = np.array(list(map(lambda x: x.days, quit_data['quit_date'] - quit_data['join_date'])))\n", "\n", "hist_kws={'histtype': 'bar', 'edgecolor':'black', 'alpha': 0.2}\n", "fig, ax = plt.subplots(figsize=(15, 6))\n", "sns.distplot(work_days, bins=100, kde=False, ax=ax, hist_kws=hist_kws)\n", "ax.set_title('Histogram of Working Days', fontsize=16)\n", "ax.set_xlabel('Working Days', fontsize=12)\n", "ax.set_ylabel('Frequency', fontsize=12)\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4gAAAGICAYAAAADL87nAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ng4AIAABAEgERAACA4f8HpFhEkO/HkoMAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Week No. for quiting\n", "quit_week = np.array(list(map(lambda x: x.month, quit_data['quit_date'])))\n", "\n", "fig, ax = plt.subplots(figsize=(15, 6))\n", "sns.countplot(quit_week, ax=ax)\n", "ax.set_title('Countplot of Quit Month', fontsize=16)\n", "ax.set_xlabel('Month', fontsize=12)\n", "ax.set_ylabel('Frequency', fontsize=12)\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4EAAAGICAYAAAD/OuUtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Week No. for quiting\n", "quit_week = np.array(list(map(lambda x: x.week, quit_data['quit_date'])))\n", "\n", "fig, ax = plt.subplots(figsize=(15, 6))\n", "sns.countplot(quit_week, ax=ax)\n", "ax.set_title('Countplot of Quit Week No.', fontsize=16)\n", "ax.set_xlabel('Week Number', fontsize=12)\n", "ax.set_ylabel('Frequency', fontsize=12)\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
employee_idcompany_iddeptsenioritysalaryjoin_datequit_datework_daysquit_week
013021.07customer_service2889000.02014-03-242015-10-3058544
1825355.07marketing20183000.02013-04-292014-04-0434014
3662910.07customer_service20115000.02012-05-142013-06-0738923
4256971.02data_science23276000.02011-10-172014-08-22104034
5509529.04data_science14165000.02012-01-302013-08-3057835
\n", "
" ], "text/plain": [ " employee_id company_id dept seniority salary join_date \\\n", "0 13021.0 7 customer_service 28 89000.0 2014-03-24 \n", "1 825355.0 7 marketing 20 183000.0 2013-04-29 \n", "3 662910.0 7 customer_service 20 115000.0 2012-05-14 \n", "4 256971.0 2 data_science 23 276000.0 2011-10-17 \n", "5 509529.0 4 data_science 14 165000.0 2012-01-30 \n", "\n", " quit_date work_days quit_week \n", "0 2015-10-30 585 44 \n", "1 2014-04-04 340 14 \n", "3 2013-06-07 389 23 \n", "4 2014-08-22 1040 34 \n", "5 2013-08-30 578 35 " ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Choose quit data\n", "quit_data['work_days'] = work_days\n", "quit_data['quit_week'] = quit_week\n", "\n", "quit_data.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Decision Tree Model" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
employee_idcompany_iddeptsenioritysalaryjoin_datequit_datequit
013021.07customer_service2889000.02014-03-242015-10-300
1825355.07marketing20183000.02013-04-292014-04-041
2927315.04marketing14101000.02014-10-13NaT0
3662910.07customer_service20115000.02012-05-142013-06-071
4256971.02data_science23276000.02011-10-172014-08-220
\n", "
" ], "text/plain": [ " employee_id company_id dept seniority salary join_date \\\n", "0 13021.0 7 customer_service 28 89000.0 2014-03-24 \n", "1 825355.0 7 marketing 20 183000.0 2013-04-29 \n", "2 927315.0 4 marketing 14 101000.0 2014-10-13 \n", "3 662910.0 7 customer_service 20 115000.0 2012-05-14 \n", "4 256971.0 2 data_science 23 276000.0 2011-10-17 \n", "\n", " quit_date quit \n", "0 2015-10-30 0 \n", "1 2014-04-04 1 \n", "2 NaT 0 \n", "3 2013-06-07 1 \n", "4 2014-08-22 0 " ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Choose the subset data\n", "stop_date = pd.to_datetime('2015-12-13') - pd.DateOffset(days=365 + 31)\n", "subset = data[data['join_date'] < stop_date]\n", "\n", "# Binary label for early quit (less than 13 months)\n", "quit = subset['quit_date'].isnull() | (subset['quit_date'] > subset['join_date'] + pd.DateOffset(days=396))\n", "subset['quit'] = 1 - quit.astype(int)\n", "\n", "subset.head()" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "# # One-hot encoding\n", "# subset['company_id'] = subset['company_id'].astype(str)\n", "# dummies = pd.get_dummies(subset[['company_id', 'dept']])\n", "# train_x = pd.concat(objs=[subset[['seniority', 'salary']], dummies], axis=1)\n", "# train_y = subset['quit'].values\n", "\n", "# train_x.head()" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
company_idsenioritysalarydept
072889000.00
1720183000.04
2414101000.04
3720115000.00
4223276000.01
\n", "
" ], "text/plain": [ " company_id seniority salary dept\n", "0 7 28 89000.0 0\n", "1 7 20 183000.0 4\n", "2 4 14 101000.0 4\n", "3 7 20 115000.0 0\n", "4 2 23 276000.0 1" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Label encoder\n", "le = LabelEncoder()\n", "train_x = subset[['company_id', 'seniority', 'salary']]\n", "train_x['dept'] = le.fit_transform(subset['dept'])\n", "train_y = subset['quit'].values\n", "\n", "train_x.head()" ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "collapsed": true }, "outputs": [], "source": [ "# Build decision tree\n", "clf = DecisionTreeClassifier(max_depth=3, min_samples_leaf=30, random_state=42)\n", "clf = clf.fit(X=train_x, y=train_y)" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "data": { "image/svg+xml": [ "\n", "\n", "\n", "\n", "\n", "\n", "Tree\n", "\n", "\n", "\n", "0\n", "\n", "salary ≤ 224500.0\n", "gini = 0.42\n", "samples = 19270\n", "value = [13480, 5790]\n", "class = Not quit\n", "\n", "\n", "\n", "1\n", "\n", "salary ≤ 62500.0\n", "gini = 0.435\n", "samples = 16506\n", "value = [11230, 5276]\n", "class = Not quit\n", "\n", "\n", "\n", "0->1\n", "\n", "\n", "True\n", "\n", "\n", "\n", "8\n", "\n", "salary ≤ 274500.0\n", "gini = 0.303\n", "samples = 2764\n", "value = [2250, 514]\n", "class = Not quit\n", "\n", "\n", "\n", "0->8\n", "\n", "\n", "False\n", "\n", "\n", "\n", "2\n", "\n", "seniority ≤ 6.5\n", "gini = 0.385\n", "samples = 2887\n", "value = [2136, 751]\n", "class = Not quit\n", "\n", "\n", "\n", "1->2\n", "\n", "\n", "\n", "\n", "\n", "5\n", "\n", "salary ≤ 182500.0\n", "gini = 0.444\n", "samples = 13619\n", "value = [9094, 4525]\n", "class = Not quit\n", "\n", "\n", "\n", "1->5\n", "\n", "\n", "\n", "\n", "\n", "3\n", "\n", "gini = 0.376\n", "samples = 2334\n", "value = [1749, 585]\n", "class = Not quit\n", "\n", "\n", "\n", "2->3\n", "\n", "\n", "\n", "\n", "\n", "4\n", "\n", "gini = 0.42\n", "samples = 553\n", "value = [387, 166]\n", "class = Not quit\n", "\n", "\n", "\n", "2->4\n", "\n", "\n", "\n", "\n", "\n", "6\n", "\n", "gini = 0.448\n", "samples = 11249\n", "value = [7443, 3806]\n", "class = Not quit\n", "\n", "\n", "\n", "5->6\n", "\n", "\n", "\n", "\n", "\n", "7\n", "\n", "gini = 0.423\n", "samples = 2370\n", "value = [1651, 719]\n", "class = Not quit\n", "\n", "\n", "\n", "5->7\n", "\n", "\n", "\n", "\n", "\n", "9\n", "\n", "seniority ≤ 17.5\n", "gini = 0.349\n", "samples = 1436\n", "value = [1113, 323]\n", "class = Not quit\n", "\n", "\n", "\n", "8->9\n", "\n", "\n", "\n", "\n", "\n", "12\n", "\n", "salary ≤ 339500.0\n", "gini = 0.246\n", "samples = 1328\n", "value = [1137, 191]\n", "class = Not quit\n", "\n", "\n", "\n", "8->12\n", "\n", "\n", "\n", "\n", "\n", "10\n", "\n", "gini = 0.381\n", "samples = 410\n", "value = [305, 105]\n", "class = Not quit\n", "\n", "\n", "\n", "9->10\n", "\n", "\n", "\n", "\n", "\n", "11\n", "\n", "gini = 0.335\n", "samples = 1026\n", "value = [808, 218]\n", "class = Not quit\n", "\n", "\n", "\n", "9->11\n", "\n", "\n", "\n", "\n", "\n", "13\n", "\n", "gini = 0.262\n", "samples = 1181\n", "value = [998, 183]\n", "class = Not quit\n", "\n", "\n", "\n", "12->13\n", "\n", "\n", "\n", "\n", "\n", "14\n", "\n", "gini = 0.103\n", "samples = 147\n", "value = [139, 8]\n", "class = Not quit\n", "\n", "\n", "\n", "12->14\n", "\n", "\n", "\n", "\n", "\n" ], "text/plain": [ "" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Visualization\n", "features = list(train_x.columns)\n", "targets = ['Not quit', 'Quit']\n", "dot_data = tree.export_graphviz(clf, out_file=None, feature_names=features, class_names=targets, \n", " filled=True, rounded=True, special_characters=True, ) \n", "graph = graphviz.Source(dot_data)\n", "graph" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "salary | 0.97439\n", "seniority | 0.02561\n", "company_id | 0.00000\n", "dept | 0.00000\n" ] } ], "source": [ "# Feature importance\n", "importance = sorted(zip(features, clf.feature_importances_), key=lambda x:x[1], reverse=True)\n", "for feature, val in importance:\n", " print('{0:10s} | {1:.5f}'.format(feature, val))" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAA54AAAGwCAYAAADBtrqIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRDqVCk8RERERERHpVP8HkeIqBw/TvZkAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Visualization\n", "hist_kws={'histtype': 'bar', 'edgecolor':'black', 'alpha': 0.2}\n", "fig, ax = plt.subplots(figsize=(15, 7))\n", "sns.distplot(subset[subset['quit']==0]['salary'], \n", " label='Not Quit', ax=ax, hist_kws=hist_kws)\n", "sns.distplot(subset[subset['quit']==1]['salary'], \n", " label='Quit', ax=ax, hist_kws=hist_kws)\n", "ax.set_xlabel('Salary', fontsize=12)\n", "ax.set_ylabel('PDF', fontsize=12)\n", "ax.legend()\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Other Factors" ] }, { "cell_type": "markdown", "metadata": { "collapsed": true }, "source": [ "1. Given how important is salary, I would definitely love to have as a variable the salary the employee who quit was offered in the next job. Otherwise, things like: promotions or raises received during the employee tenure would be interesting.\n", "\n", "2. The major findings are that employees quit at year anniversaries or at the beginning of the year. Both cases make sense. Even if you don’t like your current job, you often stay for 1 yr before quitting + you often get stocks after 1 yr so it makes sense to wait. Also, the beginning of the year is well known to be the best time to change job: companies are hiring more and you often want to stay until end of Dec to get the calendar year bonus.\n", "\n", "3. Employees with low and high salaries are less likely to quit. Probably because employees with high salaries are happy there and employees with low salaries are not that marketable, so they have a hard time finding a new job." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.5" } }, "nbformat": 4, "nbformat_minor": 2 }