diff --git a/ixda survey 2018.ipynb b/ixda survey 2018.ipynb
index 3d2363f..819ea9c 100644
--- a/ixda survey 2018.ipynb	
+++ b/ixda survey 2018.ipynb	
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 69,
    "metadata": {
     "collapsed": true
    },
@@ -22,7 +22,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 70,
    "metadata": {
     "collapsed": true
    },
@@ -34,7 +34,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 71,
    "metadata": {
     "collapsed": true
    },
@@ -48,7 +48,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 72,
    "metadata": {
     "collapsed": true
    },
@@ -78,21 +78,31 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
+   "execution_count": 73,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Survey completed by  181  people, we removed  -14  , and now we are left with 167  people\n"
+     ]
+    }
+   ],
    "source": [
     "# Cleanup - Remove positions that are not design specific\n",
     "# helper: list all distinct positions via `set(df['Q2'])`\n",
     "whitelisted_positions = ['Amaterski dizajner','CX dizajner','Dizajner interakcija','Dizajner vizualnih komunikacija','Grafički dizajner','Interactive Art Director','Marketing/Brand dizajner','Miks - Web dizajner, Dizajner interakcija, Grafički dizajner, Marketing/brand dizajner, Dizajner vizualnih komunikacijam Copywriter, UX/UI dizajner','Multimedijalni dizajner','Produkt dizajner (digitalni proizvodi)','Researcher','Student / UI dizajner / Produkt dizajner (digitalni proizvodi)','UI dizajner','UI/UX/grafički dizajner/dizajner vizualnih komunikacija','UX dizajner','UX/UI dizajner','Web dizajner','Web dizajner, graficki dizajner, marketing/branding dizajner, ux/ui dizajner']\n",
-    "df = df[df['Q2'].isin(whitelisted_positions)]"
+    "rows_before_cleanup = len(df.index)\n",
+    "df = df[df['Q2'].isin(whitelisted_positions)]\n",
+    "rows_after_cleanup = len(df.index)\n",
+    "\n",
+    "print('Survey completed by ', rows_before_cleanup, ' people, we removed ', (rows_before_cleanup - rows_after_cleanup), ' inputs, and now we are left with', rows_after_cleanup, ' people')"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 74,
    "metadata": {
     "collapsed": true
    },
@@ -120,7 +130,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 75,
    "metadata": {},
    "outputs": [
     {
@@ -144,7 +154,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 76,
    "metadata": {},
    "outputs": [
     {
@@ -178,10 +188,10 @@
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x109313ef0>"
+       "<matplotlib.axes._subplots.AxesSubplot at 0x1a12d9a048>"
       ]
      },
-     "execution_count": 10,
+     "execution_count": 76,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -189,7 +199,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAG6CAYAAAAYgb2XAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x104b0ca20>"
+       "<matplotlib.figure.Figure at 0x1a1347bb38>"
       ]
      },
      "metadata": {},
@@ -207,7 +217,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 77,
    "metadata": {
     "collapsed": true
    },
@@ -222,7 +232,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 78,
    "metadata": {},
    "outputs": [
     {
@@ -373,7 +383,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 79,
    "metadata": {},
    "outputs": [
     {
@@ -487,10 +497,10 @@
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x103dc8630>"
+       "<matplotlib.axes._subplots.AxesSubplot at 0x1a13a022e8>"
       ]
      },
-     "execution_count": 13,
+     "execution_count": 79,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -498,7 +508,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD4CAYAAAD8Zh1EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAFrtJREFUeJzt3X2QXHWd7/H3N08EQlyQjBQSZJAb\nkYSBQfKgoiAbiMGrEPD6gBeNKIZSoyAWD6V/mGvp1cXdpVDXRfaCBs0ql0BKvEWxIia7gkAyCYEk\nBAjkgo6EZEhWeUyGhO/9oztzJ2SG6Uw/TOfk/ao6NadPn+7ft2d6PnPmd37n15GZSJL2fsOGugBJ\nUm0Y6JJUEAa6JBWEgS5JBWGgS1JBGOiSVBAGuiQVhIEuSQVhoEtSQYxoZGPjxo3L1tbWRjYpSXu9\n5cuXP5uZLQPt19BAb21tpaOjo5FNStJeLyKeqmQ/u1wkqSAMdEkqCANdkgqioX3okhrjlVdeobOz\nk61btw51KdoDo0ePZvz48YwcOXJQjzfQpQLq7Oxk7NixtLa2EhFDXY4qkJls3ryZzs5OjjrqqEE9\nx4BdLhExOiKWRsSDEbEmIv5HeftREXF/RKyLiJsiYtSgKpBUc1u3buWQQw4xzPciEcEhhxxS1X9V\nlfShbwP+NjNPANqBmRHxTuDvgKszcwLwn8BnB12FpJozzPc+1f7MBgz0LHmhfHNkeUngb4GF5e3z\ngVlVVSJJqkpFo1wiYnhErAQ2AXcCTwB/yczt5V06gcP7eeyciOiIiI6urq5a1KydYoBFKhvorbKn\ni5pTRYGemTsysx0YD0wFju1rt34ee11mTs7MyS0tA165KqlAOjs7Ofvss5kwYQJvfetbmTt3Ltu2\nbWPp0qW0t7fT3t7OCSecwKJFi3oe8/LLL3PqqaeyY8eOPWqrtbWVZ599lu7ubk455RS2b98+8IMG\n6ac//Slz586t2/MP1h6NQ8/MvwBLgHcCB0XEzlEy44Gna1uapL1ZZnLuuecya9Ys1q1bx7p163j5\n5Ze5/PLLOe644+jo6GDlypXccccdXHTRRT0BfMMNN3DuuecyfPjwXZ6v0oAfNWoU06dP56abbqr5\na2p2lYxyaYmIg8rr+wOnA2uBxcB/K+82G/hVvYqUtPf53e9+x+jRo7ngggsAGD58OFdffTU33ngj\nr776KiNGlI4Ht27dusvJwAULFnD22WcDsGTJEk477TQ+8YlP0NbWBsDPf/5zpk6dSnt7OxdddFGf\nQT9r1iwWLFjQb21Llizh1FNP5aMf/Shve9vbuPLKK1mwYAFTp06lra2NJ554AoBf//rXTJs2jRNP\nPJHTTz+djRs37vZcXV1dfPjDH2bKlClMmTKFe+65Z5DfsepVcoR+GLA4Ih4ClgF3Zub/Aa4ALo2I\nx4FDgOvrV6akvc2aNWs46aSTdtn2hje8gdbWVh5//HHuv/9+Jk2aRFtbG9deey0jRoygu7ub9evX\n03tW1qVLl/Ltb3+bhx9+mLVr13LTTTdxzz33sHLlSoYPH95ncB933HEsW7bsdet78MEHueaaa1i1\nahU/+9nPeOyxx1i6dCkXXnghP/jBDwB4z3vew3333ccDDzzAxz/+ca666qrdnufiiy/mK1/5CsuW\nLeOWW27hwgsvHMR3qzYGvLAoMx8CTuxj+3pK/emStJvM7HMYXmbpdNu0adNYs2YNa9euZfbs2Zx5\n5pls2bKFgw46aJf9p06d2nOhzV133cXy5cuZMmUKUOpvf9Ob3rRbG8OHD2fUqFE8//zzjB07ts/6\npkyZwmGHHQbA0UcfzYwZMwBoa2tj8eLFQOkcwMc+9jE2bNhAd3d3nxf8/Pa3v+Xhhx/uuf3cc8+9\nbrv15JWikupi0qRJ3HLLLbtse+6559i4cSPHHHNMz7Zjjz2WMWPGsHr1ao4++ujdLqwZM2ZMz3pm\nMnv2bL7zne8M2P62bdsYPXp0v/fvt99+PevDhg3ruT1s2LCe/vwvfelLXHrppZx11lksWbKEefPm\n7fY8r776Kvfeey/777//gDXVm5NzSfuArPFSienTp/PSSy9x4403AqWTml/96leZO3cuzzzzTE9o\nPvXUUzz66KO0trZy8MEHs2PHjn6vlpw+fToLFy5k06ZNAGzZsoWnntp9qvDNmzfT0tIy6DlRdvrr\nX//K4YeXRmTPnz+/z31mzJjBD3/4w57bK1eurKrNahjokuoiIli0aBELFy5kwoQJHHLIIQwbNoyv\nf/3r3H333Zxwwgm0t7dzzjnn8KMf/Yhx48YBpYC8++67+3zOiRMn8q1vfYsZM2Zw/PHHc8YZZ7Bh\nw4bd9lu8eDEf+MAHqn4N8+bN4yMf+Qjvfe97e+p7re9///t0dHRw/PHHM3HiRK699tqq2x20zGzY\nctJJJ6VqaKDvuPZZDz/88FCXsJt77rkn3/KWt2RHR8fr7rdixYo8//zzq2rrnHPOyUceeaSq5xgq\nff3sgI6sIGPtQ5fUEO9+97v77B55rRNPPJHTTjuNHTt27DYWvRLd3d3MmjVrl376fYWBLqnpfOYz\nnxn0Y0eNGsWnPvUpAFatWsUnP/nJXe7fb7/9uP/++6uqr1kZ6JIKq62tbUhPUjaaJ0UlqSAMdEkq\nCANdkgrCPnRpX1DrScwrvbpIDeURuqS6aeb50C+77DImTZrEZZddNqjXVolPf/rTLFy4cOAda8RA\nl1QX2eTzof/4xz9mxYoVfO973xvcC2xCBrqkumjm+dDPOussXnzxRaZNm8ZNN93U75zm8+bNY/bs\n2cyYMYPW1lZuvfVWLr/8ctra2pg5cyavvPIKAN/85jeZMmUKxx13HHPmzOmZUbK35cuXc+qpp3LS\nSSfx/ve/v88pC6pWyeWktVq89L/GvPRf/djt8vFa/zZX4JprrslLLrlkt+3t7e35wAMP5H333ZcT\nJ07MMWPG5K233pqZmdu2bctDDz20Z9/FixfnAQcckOvXr+95XR/84Aezu7s7MzM///nP5/z58zMz\n88gjj8yurq7MzNy+fXuOGzfudesbM2ZMz/p5552Xv//97zMz86mnnsq3v/3tmZn5jW98I08++eTs\n7u7OlStX5v7775+33357ZmbOmjUrFy1alJmZmzdv7nmu888/P2+77bbMzJw9e3befPPN2d3dne96\n17ty06ZNmZn5y1/+Mi+44II+6/LSf0lNJ5t8PvTe+pvTHODMM89k5MiRtLW1sWPHDmbOnAmULlp6\n8skngdJkYFdddRUvvfQSW7ZsYdKkSXzoQx/qeb5HH32U1atXc8YZZwCl7qOdc7HXkoEuqS6afT70\n3l5vTvPe86SPHDmy54/UznnTt27dyhe+8AU6Ojo44ogjmDdv3m6vITOZNGkS9957b0X1DJZ96NK+\nYAgmRN+b5kOvZk7znbWOGzeOF154oc9RLccccwxdXV09gf7KK6+wZs2aituolIEuqS72pvnQq5nT\n/KCDDuJzn/scbW1tzJo1q6c7qLdRo0axcOFCrrjiip7X/Yc//KHiNipWSUd7rRZPitaYJ0XVD+dD\ndz50Saob50OvPwNdUtNxPvTBMdClgsp+hg3uS/a2+dCzjwuS9oQnRaUCGj16NJs3b646INQ4mcnm\nzZsrHmrZF4/QpQIaP348nZ2ddHV1DXUp2gOjR49m/Pjxg368gS4V0MiRI3uurtS+Y8Aul4g4IiIW\nR8TaiFgTEReXt8+LiD9HxMryUvmgT0lSzVVyhL4d+GpmroiIscDyiLizfN/Vmfn39StPklSpAQM9\nMzcAG8rrz0fEWuDwehcmSdozezTKJSJagROBnYM450bEQxFxQ0Qc3M9j5kRER0R0eIJGkuqn4kCP\niAOBW4BLMvM54J+Bo4F2Skfw/9DX4zLzusycnJmTW1paalCyJKkvFQV6RIykFOYLMvNWgMzcmJk7\nMvNV4F+AqfUrU5I0kEpGuQRwPbA2M/+x1/bes7OfA6yufXmSpEpVMsrlZOCTwKqI2HkN7deA8yKi\nndLsyE8CF9WlQklSRSoZ5XI30NeEELfXvhxJ0mA5l4skFYSBLkkFYaBLUkEY6JJUEAa6JBWEgS5J\nBWGgS1JBGOiSVBAGuiQVhIEuSQVhoEtSQRjoklQQBrokFYSBLkkFYaBLUkFU8gEX2of1NRF+b9mQ\nKiRVwiN0SSoIA12SCsJAl6SCMNAlqSAMdEkqCANdkgrCQJekgjDQJakgDHRJKogBAz0ijoiIxRGx\nNiLWRMTF5e1vjIg7I2Jd+evB9S9XktSfSo7QtwNfzcxjgXcCX4yIicCVwF2ZOQG4q3xbkjREBgz0\nzNyQmSvK688Da4HDgbOB+eXd5gOz6lWkJGlge9SHHhGtwInA/cChmbkBSqEPvKmfx8yJiI6I6Ojq\n6qquWjWfqGCR1BAVB3pEHAjcAlySmc9V+rjMvC4zJ2fm5JaWlsHUKEmqQEWBHhEjKYX5gsy8tbx5\nY0QcVr7/MGBTfUqUJFWiklEuAVwPrM3Mf+x1123A7PL6bOBXtS9PklSpSj7g4mTgk8CqiFhZ3vY1\n4LvA/46IzwJ/BD5SnxIlSZUYMNAz8276P7U1vbblSJIGyytFJakg/EzRJlXJaD8/z1NSbx6hS1JB\nGOiSVBAGuiQVhIEuSQVhoEtSQRjoklQQBrokFYSBLkkFYaBLUkEY6JJUEAa6JBWEgS5JBWGgS1JB\nGOiSVBAGuiQVhIEuSQVhoEtSQRjoklQQBrokFYSBLkkFYaBLUkEY6JJUEAa6JBWEgS5JBTFgoEfE\nDRGxKSJW99o2LyL+HBEry8sH6lumJGkglRyh/xSY2cf2qzOzvbzcXtuypOYTAyzSUBsw0DPzP4At\nDahFklSFavrQ50bEQ+UumYP72yki5kRER0R0dHV1VdGcJOn1DDbQ/xk4GmgHNgD/0N+OmXldZk7O\nzMktLS2DbE6SNJBBBXpmbszMHZn5KvAvwNTaliVJ2lODCvSIOKzXzXOA1f3tK0lqjBED7RARvwDe\nB4yLiE7gG8D7IqIdSOBJ4KI61ihJqsCAgZ6Z5/Wx+fo61CJJqoJXikpSQRjoklQQBrokFYSBLkkF\nYaBLUkEMOMpFfahkJqasexWStAuP0CWpIAx0SSoIA12SCsJAl6SCMNAlqSAMdEkqCIctStorOXp4\ndx6hS1JBGOiSVBAGuiQVhIEuSQVhoEtSQRjoklQQBrokFYSBLkkFYaBLUkEY6JJUEAa6JBWEgS5J\nBTFgoEfEDRGxKSJW99r2xoi4MyLWlb8eXN8yJUkDqeQI/afAzNdsuxK4KzMnAHeVb0uShtCAgZ6Z\n/wFsec3ms4H55fX5wKwa1yVJ2kOD7UM/NDM3AJS/vqm/HSNiTkR0RERHV1fXIJuTpL1UDLDUUN1P\nimbmdZk5OTMnt7S01Ls5SdpnDTbQN0bEYQDlr5tqV5IkaTAGG+i3AbPL67OBX9WmHEnSYFUybPEX\nwL3AMRHRGRGfBb4LnBER64AzyrclSUNowA+Jzszz+rlreo1rkSRVwStFJakgDHRJKggDXZIKwkCX\npIIw0CWpIAx0SSqIAYctSnuFgebEyH2kBu3TPEKXpIIw0CWpIAx0SSoIA12SCsKTon3w3JaaVSWf\nh+D7c9/lEbokFYSBLkkFYaBLUkEY6JJUEAa6JBWEo1wk7TFHgjUnj9AlqSAMdEkqCANdkgrCQJek\ngjDQJakgDHRJKgiHLUpSFZppCKdH6JJUEFUdoUfEk8DzwA5ge2ZOrkVRkqQ9V4sul9My89kaPI8k\nqQp2uUhSQVQb6An8JiKWR8ScvnaIiDkR0RERHV1dXVU2J2lAMcCiwqo20E/OzHcAZwJfjIhTXrtD\nZl6XmZMzc3JLS0uVzUmS+lNVoGfm0+Wvm4BFwNRaFCVJ2nODDvSIGBMRY3euAzOA1bUqTJK0Z6oZ\n5XIosCgidj7Pv2bmHTWpSpK0xwYd6Jm5HjihhrVIkqrgsEVJKggDXZIKwsm51PQqGTrtZ1hKHqFL\nUmEY6JJUEHa5SCquZpqsvAE8QpekgjDQJakgDHRJKggDXZIKwkCXpIIw0CWpIAx0SSoIA12SCsJA\nl6SCMNAlqSAMdEkqiL1vLpd9bG4GSaqUR+iSVBAGuiQVhIEuSQVhoEtSQRjoklQQBrokFYSBLkkF\nUVWgR8TMiHg0Ih6PiCtrVZQkac8NOtAjYjjwT8CZwETgvIiYWKvCJEl7ppoj9KnA45m5PjO7gV8C\nZ9emLEnSnqrm0v/DgT/1ut0JTHvtThExB5hTvvlCRDxaRZsA44Bn+713oKkBaiAGqqFZ6miGGpql\njn2khmapo6L3RQPsFd+Lymo4spKdqgn0vsrYbSaVzLwOuK6KdnZtNKIjMyfX6vn21hqapY5mqKFZ\n6miGGpqljmaooVnqaGQN1XS5dAJH9Lo9Hni6unIkSYNVTaAvAyZExFERMQr4OHBbbcqSJO2pQXe5\nZOb2iJgL/BswHLghM9fUrLL+1az7pgrNUAM0Rx3NUAM0Rx3NUAM0Rx3NUAM0Rx0NqyEynUBckorA\nK0UlqSAMdEkqCANdkgqiqT9TNCLeTunq08MpjXF/GrgtM9cOaWH7sIiYCmRmLitP9TATeCQzbx/C\nmm7MzE8NVfvSTr1G/D2dmb+NiE8A7wbWAtdl5it1bb9ZT4pGxBXAeZSmFOgsbx5P6Zv1y8z87lDV\nNhTKf9wOB+7PzBd6bZ+ZmXc0qIZvUJq7ZwRwJ6Urg5cApwP/lpnfbkANrx0aG8BpwO8AMvOsetfQ\nl4h4D6XpMFZn5m8a1OY0YG1mPhcR+wNXAu8AHgb+Z2b+tRF1lGs5GjiH0rUp24F1wC8aVUNEfBlY\nlJl/GnDn+taxgNLvxwHAX4ADgVuB6ZTydnZd22/iQH8MmPTav2jlv4BrMnPC0FS2Sy0XZOZPGtDO\nl4EvUvor3w5cnJm/Kt+3IjPfUe8aym2tKre/H/AMML5XmNyfmcc3oIYVlALrf1H6ry2AX1D6Q09m\n/nu9ayjXsTQzp5bXP0fp57MImAH8uhEHHBGxBjihPIT4OuAlYCGl8DghM8+tdw3lOr4MfAj4d+AD\nwErgPykF/Bcyc0kDavgr8CLwBKX3w82Z2VXvdvuo46HMPD4iRgB/Bt6cmTsiIoAH6/47kplNuQCP\nAEf2sf1I4NGhrq9cyx8b1M4q4MDyeivQQSnUAR5o4Ot9oK/18u2VDaphGPAVSv8htJe3rR+Cn33v\n78UyoKW8PgZY1aAa1vZaXzEUP49yW6uA4eX1A4Al5fW3NOr9CTxQfm/MAK4HuoA7gNnA2AZ+L1YD\no4CDgeeBN5a3j+7986rX0sx96JcAd0XEOv7/JGBvAf4LMLdRRUTEQ/3dBRzaoDKGZ7mbJTOfjIj3\nAQsj4kgaNuUTAN0RcUBmvgSctHNjRPwN8GojCsjMV4GrI+Lm8teNDM25oGERcTClEIksHw1m5osR\nsb1BNazu9V/igxExOTM7IuJtQF37avswAthB6b+3sQCZ+ceIGNmg9rP83vgN8Jtyu2dS6rb9e6Cl\nQXVcT+lgdDjwdeDmiFgPvJNS93FdNW2XC0BEDKPUL3k4peDqBJZl5o4G1rAReD+lfyF3uQv4Q2a+\nuQE1/A64NDNX9to2ArgB+O+ZObzeNZTb3C8zt/WxfRxwWGauakQdr2n7vwInZ+bXGtzuk5T+iAWl\nrp93Z+YzEXEgcHdmtjeghr8BrgHeS2k2v3dQOvj5E/DlzHyw3jWU67gY+CxwH3AK8HeZ+ZOIaAFu\nycxTGlDDA5l5Yj/37Z+ZL9e7hl7tvRkgM5+OiIMonWP6Y2YurXvbzRzozSAirgd+kpl393Hfv2bm\nJxpQw3hge2Y+08d9J2fmPfWuQZWJiAOAQzPz/zawzbHAWykdJXdm5sZGtd2rhknAsZROCj8yBO2/\nLTMfa3S7zcZAl6SC8MIiSSoIA12SCsJAl6SCMNAlqSD+H9JpZ2OWJZmbAAAAAElFTkSuQmCC\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x109338fd0>"
+       "<matplotlib.figure.Figure at 0x1a1380a748>"
       ]
      },
      "metadata": {},
@@ -513,7 +523,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 80,
    "metadata": {},
    "outputs": [
     {
@@ -522,7 +532,7 @@
        "RanksumsResult(statistic=-2.4321521980337617, pvalue=0.015009398838606273)"
       ]
      },
-     "execution_count": 14,
+     "execution_count": 80,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -543,7 +553,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 81,
    "metadata": {},
    "outputs": [
     {
@@ -604,7 +614,7 @@
        "18            ZAGREBAČKA         25"
       ]
      },
-     "execution_count": 15,
+     "execution_count": 81,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -626,7 +636,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 82,
    "metadata": {
     "collapsed": true
    },
@@ -644,7 +654,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 83,
    "metadata": {},
    "outputs": [
     {
@@ -759,7 +769,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAGfCAYAAABP4CkSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nER2Rgh4R0REp6BERHfH/AaM3qANwhba1AAAAAElFTkSuQmCC\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x10965da90>"
+       "<matplotlib.figure.Figure at 0x1a1354d748>"
       ]
      },
      "metadata": {},
@@ -781,7 +791,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 84,
    "metadata": {},
    "outputs": [
     {
@@ -868,7 +878,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAGfCAYAAAB/bDgzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npMYY3JLUmP8PHoomC/Vi8fIAAAAASUVORK5CYII=\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x109873e10>"
+       "<matplotlib.figure.Figure at 0x1a1355f2e8>"
       ]
      },
      "metadata": {},
@@ -888,7 +898,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 85,
    "metadata": {},
    "outputs": [
     {
@@ -982,7 +992,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAGfCAYAAAB/bDgzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nsbglqTMWtyR1xuKWpM78f3Vq0yckgzeoAAAAAElFTkSuQmCC\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x1095da3c8>"
+       "<matplotlib.figure.Figure at 0x1a149f3668>"
       ]
      },
      "metadata": {},
@@ -1002,7 +1012,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 86,
    "metadata": {},
    "outputs": [
     {
@@ -1096,7 +1106,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAF+CAYAAACidPAUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nUwAAAABJRU5ErkJggg==\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x103dc1240>"
+       "<matplotlib.figure.Figure at 0x1a14a400f0>"
       ]
      },
      "metadata": {},
@@ -1116,7 +1126,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 87,
    "metadata": {
     "collapsed": true
    },
@@ -1132,7 +1142,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 88,
    "metadata": {
     "collapsed": true
    },
@@ -1152,7 +1162,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 89,
    "metadata": {},
    "outputs": [
     {
@@ -1237,10 +1247,10 @@
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x10ebb01d0>"
+       "<matplotlib.axes._subplots.AxesSubplot at 0x1a14d670b8>"
       ]
      },
-     "execution_count": 23,
+     "execution_count": 89,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -1248,7 +1258,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEdCAYAAAAcmJzBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nlggXuplZIv4//iCFyi7Jh+4AAAAASUVORK5CYII=\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x10e9ce550>"
+       "<matplotlib.figure.Figure at 0x1a14c58a90>"
       ]
      },
      "metadata": {},
@@ -1270,7 +1280,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": 90,
    "metadata": {},
    "outputs": [
     {
@@ -1356,10 +1366,10 @@
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x10ea87080>"
+       "<matplotlib.axes._subplots.AxesSubplot at 0x1a14e2f198>"
       ]
      },
-     "execution_count": 24,
+     "execution_count": 90,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -1367,7 +1377,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAFdCAYAAAATwJatAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEVHE/wd3TdVCHsKJZgAAAABJRU5ErkJggg==\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x1098deba8>"
+       "<matplotlib.figure.Figure at 0x1a14a38da0>"
       ]
      },
      "metadata": {},
@@ -1391,7 +1401,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 91,
    "metadata": {},
    "outputs": [
     {
@@ -1463,10 +1473,10 @@
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x10ebb0ac8>"
+       "<matplotlib.axes._subplots.AxesSubplot at 0x1a14c6b748>"
       ]
      },
-     "execution_count": 25,
+     "execution_count": 91,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -1474,7 +1484,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAFdCAYAAAATwJatAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIyx0SWqEhS5JjbDQJakR/x8lODTiJclA+QAAAABJRU5ErkJggg==\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x10ebb9390>"
+       "<matplotlib.figure.Figure at 0x1a14e6ca20>"
       ]
      },
      "metadata": {},
@@ -1505,7 +1515,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 92,
    "metadata": {},
    "outputs": [
     {
@@ -1562,7 +1572,7 @@
        "Name: Q1, dtype: int64"
       ]
      },
-     "execution_count": 26,
+     "execution_count": 92,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1575,16 +1585,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 93,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x1a12ad9208>"
+       "<matplotlib.axes._subplots.AxesSubplot at 0x1a14f12b70>"
       ]
      },
-     "execution_count": 27,
+     "execution_count": 93,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -1592,7 +1602,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAABBIAAAKTCAYAAACkZCuBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrkJggg==\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x1a12ad3d30>"
+       "<matplotlib.figure.Figure at 0x1a1392ebe0>"
       ]
      },
      "metadata": {},
@@ -1615,7 +1625,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 94,
    "metadata": {},
    "outputs": [
     {
@@ -1672,7 +1682,7 @@
        "Name: Q1, dtype: int64"
       ]
      },
-     "execution_count": 28,
+     "execution_count": 94,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1684,16 +1694,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 95,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x1a12b116d8>"
+       "<matplotlib.axes._subplots.AxesSubplot at 0x1a15123be0>"
       ]
      },
-     "execution_count": 29,
+     "execution_count": 95,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -1701,7 +1711,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAABBIAAAH2CAYAAADNta/1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x1a12ca6a90>"
+       "<matplotlib.figure.Figure at 0x1a1392ea58>"
       ]
      },
      "metadata": {},
@@ -1732,7 +1742,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 96,
    "metadata": {},
    "outputs": [
     {
@@ -1805,10 +1815,10 @@
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x1095da6a0>"
+       "<matplotlib.axes._subplots.AxesSubplot at 0x1a15276ba8>"
       ]
      },
-     "execution_count": 30,
+     "execution_count": 96,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -1816,7 +1826,7 @@
      "data": {
       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEGCAYAAAB1iW6ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x1094dcbe0>"
+       "<matplotlib.figure.Figure at 0x1a15210e80>"
       ]
      },
      "metadata": {},
@@ -1831,7 +1841,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 97,
    "metadata": {},
    "outputs": [
     {
@@ -1852,13 +1862,114 @@
    ]
   },
   {
-   "cell_type": "code",
-   "execution_count": null,
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## By education"
+   ]
+  },
+  {
+   "cell_type": "markdown",
    "metadata": {
     "collapsed": true
    },
-   "outputs": [],
-   "source": []
+   "source": [
+    "Q9: Na skali od 1-7, koliko vas je obrazovanje pripremilo za trenutnu radnu poziciju?  \n",
+    "Q7: Obrazovanje    "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 116,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style>\n",
+       "    .dataframe thead tr:only-child th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: left;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Q7</th>\n",
+       "      <th>count</th>\n",
+       "      <th>mean</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>24</th>\n",
+       "      <td>GRF</td>\n",
+       "      <td>44</td>\n",
+       "      <td>2.727273</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>32</th>\n",
+       "      <td>Srednja stručna sprema</td>\n",
+       "      <td>40</td>\n",
+       "      <td>1.925000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>Arhitektonski fakultet - studij dizajna</td>\n",
+       "      <td>18</td>\n",
+       "      <td>3.166667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>37</th>\n",
+       "      <td>TVZ</td>\n",
+       "      <td>13</td>\n",
+       "      <td>2.769231</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>FER</td>\n",
+       "      <td>5</td>\n",
+       "      <td>3.200000</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                         Q7  count      mean\n",
+       "24                                      GRF     44  2.727273\n",
+       "32                   Srednja stručna sprema     40  1.925000\n",
+       "7   Arhitektonski fakultet - studij dizajna     18  3.166667\n",
+       "37                                      TVZ     13  2.769231\n",
+       "9                                       FER      5  3.200000"
+      ]
+     },
+     "execution_count": 116,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# count by education\n",
+    "# by_education = df.groupby('Q7').count().iloc[:,1].reset_index(name='Q7(count)')\n",
+    "# by_education.sort_values(by=['Q7(count)'], inplace=True, ascending=False)\n",
+    "\n",
+    "by_education = df.groupby('Q7')['Q9'].agg(['count', 'mean']).reset_index()\n",
+    "by_education.sort_values(by=['count'], inplace=True, ascending=False)\n",
+    "\n",
+    "# if we remove entries with less than 5 answers\n",
+    "by_education = by_education.loc[by_education['count'] >= 5]\n",
+    "by_education"
+   ]
   },
   {
    "cell_type": "code",