Group 33.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a21a8a2b",
   "metadata": {},
   "source": [
    "## 1. Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9228af4f",
   "metadata": {},
   "source": [
    "#### 1.1 Background Information"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dd254685",
   "metadata": {},
   "source": [
    "According to Forbes, approximately 66% of users between the age of 18 and 35 years old shows stronger loyalty to brands with strong social media presence (Shaw, 2018). Since they are whose with highest buying power, it is crucial for companies to understand how to engage their customers effectively on social media (Sheerin et al., 2020). Specificly, we will look into key marketing metrics on Facebook platform, **determine which is the best proxy for consumer interaction and find out what type of content is the most effective to engage users.**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ac0afb6f",
   "metadata": {},
   "source": [
    "#### 1.2 Dataset Description"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5441477b",
   "metadata": {},
   "source": [
    "The dataset is a compilation of 500 different facebook posts by a cosmetics brand. Each observation includes important information on the post including the following variables of interest for this report:\n",
    "\n",
    "- **Type**: The type of facebook post (Video, Photo, Status and Link).\n",
    "- **Lifetime Impressions**: The number of times a post is displayed on Facebook users' personal feed. The post does not have to be clicked to count as an impression.\n",
    "- **Lifetime Consumptions**: The total number of clicks anywhere on the post. Each click, even on the same post, is counted as one."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b0fd16c-62e7-4652-b689-d3e5328243bb",
   "metadata": {},
   "source": [
    "#### 1.3 Project Objective"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "980f8ad1-a24e-48ec-8131-719cf15ec608",
   "metadata": {},
   "source": [
    "We decide that the Lifetime Consumptions/Lifetime Impressions ratio will be the best proxy for user engagement. This metric essentially tells you the interaction rate of a consumer once they see the content on their feed.\n",
    "\n",
    "Our goal is to find **which type of Facebook content delivers the highest Consumptions/Impressions ratio and is there enough statistical evidence to pinpoint which type does the best job?**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0985bbaf",
   "metadata": {
    "tags": []
   },
   "source": [
    "## 2. Preliminary Results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb70338b",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### 2.1 Reading the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "fa64e184",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[1mRows: \u001b[22m\u001b[34m500\u001b[39m \u001b[1mColumns: \u001b[22m\u001b[34m19\u001b[39m\n",
      "\n",
      "\u001b[36m──\u001b[39m \u001b[1mColumn specification\u001b[22m \u001b[36m────────────────────────────────────────────────────────\u001b[39m\n",
      "\u001b[1mDelimiter:\u001b[22m \";\"\n",
      "\u001b[31mchr\u001b[39m  (1): Type\n",
      "\u001b[32mdbl\u001b[39m (18): Page total likes, Category, Post Month, Post Weekday, Post Hour, P...\n",
      "\n",
      "\n",
      "\u001b[36mℹ\u001b[39m Use \u001b[30m\u001b[47m\u001b[30m\u001b[47m`spec()`\u001b[47m\u001b[30m\u001b[49m\u001b[39m to retrieve the full column specification for this data.\n",
      "\u001b[36mℹ\u001b[39m Specify the column types or set \u001b[30m\u001b[47m\u001b[30m\u001b[47m`show_col_types = FALSE`\u001b[47m\u001b[30m\u001b[49m\u001b[39m to quiet this message.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Load the required libraries\n",
    "library(tidyverse)\n",
    "library(cowplot)\n",
    "library(digest)\n",
    "library(gridExtra)\n",
    "library(infer)\n",
    "library(repr)\n",
    "library(tidyverse)\n",
    "set.seed(1)\n",
    "\n",
    "# Here we read the data set straight from the web\n",
    "# The original source is linked here (https://archive.ics.uci.edu/ml/datasets/Facebook+metrics)\n",
    "temp <- tempfile()\n",
    "download.file(\"https://archive.ics.uci.edu/ml/machine-learning-databases/00368/Facebook_metrics.zip\",temp)\n",
    "data <- read_delim(unz(temp, \"dataset_Facebook.csv\"), delim=\";\")\n",
    "unlink(temp)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a9883037",
   "metadata": {},
   "source": [
    "#### 2.2 Cleaning the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "08533e05",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"dataframe\">\n",
       "<caption>A tibble: 6 × 4</caption>\n",
       "<thead>\n",
       "\t<tr><th scope=col>Type</th><th scope=col>LifetimeImpressions</th><th scope=col>LifetimeConsumptions</th><th scope=col>ConsumptionsOverImpressions</th></tr>\n",
       "\t<tr><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "\t<tr><td>Photo </td><td> 5091</td><td> 159</td><td>0.03123159</td></tr>\n",
       "\t<tr><td>Status</td><td>19057</td><td>1674</td><td>0.08784174</td></tr>\n",
       "\t<tr><td>Photo </td><td> 4373</td><td> 154</td><td>0.03521610</td></tr>\n",
       "\t<tr><td>Photo </td><td>87991</td><td>1119</td><td>0.01271721</td></tr>\n",
       "\t<tr><td>Photo </td><td>13594</td><td> 580</td><td>0.04266588</td></tr>\n",
       "\t<tr><td>Status</td><td>20849</td><td>1389</td><td>0.06662190</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "A tibble: 6 × 4\n",
       "\\begin{tabular}{llll}\n",
       " Type & LifetimeImpressions & LifetimeConsumptions & ConsumptionsOverImpressions\\\\\n",
       " <chr> & <dbl> & <dbl> & <dbl>\\\\\n",
       "\\hline\n",
       "\t Photo  &  5091 &  159 & 0.03123159\\\\\n",
       "\t Status & 19057 & 1674 & 0.08784174\\\\\n",
       "\t Photo  &  4373 &  154 & 0.03521610\\\\\n",
       "\t Photo  & 87991 & 1119 & 0.01271721\\\\\n",
       "\t Photo  & 13594 &  580 & 0.04266588\\\\\n",
       "\t Status & 20849 & 1389 & 0.06662190\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "A tibble: 6 × 4\n",
       "\n",
       "| Type &lt;chr&gt; | LifetimeImpressions &lt;dbl&gt; | LifetimeConsumptions &lt;dbl&gt; | ConsumptionsOverImpressions &lt;dbl&gt; |\n",
       "|---|---|---|---|\n",
       "| Photo  |  5091 |  159 | 0.03123159 |\n",
       "| Status | 19057 | 1674 | 0.08784174 |\n",
       "| Photo  |  4373 |  154 | 0.03521610 |\n",
       "| Photo  | 87991 | 1119 | 0.01271721 |\n",
       "| Photo  | 13594 |  580 | 0.04266588 |\n",
       "| Status | 20849 | 1389 | 0.06662190 |\n",
       "\n"
      ],
      "text/plain": [
       "  Type   LifetimeImpressions LifetimeConsumptions ConsumptionsOverImpressions\n",
       "1 Photo   5091                159                 0.03123159                 \n",
       "2 Status 19057               1674                 0.08784174                 \n",
       "3 Photo   4373                154                 0.03521610                 \n",
       "4 Photo  87991               1119                 0.01271721                 \n",
       "5 Photo  13594                580                 0.04266588                 \n",
       "6 Status 20849               1389                 0.06662190                 "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# First remove spaces from the column names\n",
    "names(data)<-make.names(names(data),unique = TRUE)\n",
    "\n",
    "# Select only the variables of interest and calculate the ratio of consumptions/impressions\n",
    "data_selected = data %>%\n",
    "    select(Type, Lifetime.Post.Total.Impressions, Lifetime.Post.Consumptions) %>%\n",
    "    mutate(ratio=Lifetime.Post.Consumptions/Lifetime.Post.Total.Impressions) %>%\n",
    "    filter(!is.na(Type))\n",
    "\n",
    "# Assign new more manageable column names\n",
    "names(data_selected) <- c(\"Type\", \"LifetimeImpressions\", \"LifetimeConsumptions\", \"ConsumptionsOverImpressions\")\n",
    "\n",
    "# Here we break the dataset apart by type for later use\n",
    "data_photo = data_selected %>%\n",
    "    filter(Type == \"Photo\")\n",
    "data_status = data_selected %>%\n",
    "    filter(Type == \"Status\")\n",
    "data_link = data_selected %>%\n",
    "    filter(Type == \"Link\")\n",
    "data_video = data_selected %>%\n",
    "    filter(Type == \"Video\")\n",
    "\n",
    "# Preview the clean dataset\n",
    "head(data_selected)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "813d1f12",
   "metadata": {},
   "source": [
    "#### 2.3 Data Summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "18d661b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Here we will provide a summary of the dataset including means, minnimums, and maximums of the variables of interes\n",
    "data_summary <- data_selected %>%\n",
    "    group_by(Type) %>%\n",
    "    summarise(MeanImpressions = mean(LifetimeImpressions), MinImpressions = min(LifetimeImpressions),\n",
    "              MaxImpressions = max(LifetimeImpressions),  MeanConsumptions = mean(LifetimeConsumptions),\n",
    "              MinConsumptions = min(LifetimeConsumptions), MaxConsumptions = max(LifetimeConsumptions), \n",
    "              MeanRatio = mean(ConsumptionsOverImpressions), Num = sum(Type == Type))\n",
    "data_summary\n",
    "\n",
    "#Select MeanImpressions and MeanConsumptions and put them in the same column for data_summary_plot\n",
    "data_summary_longer <- data_summary %>%\n",
    "    select(Type, MeanImpressions, MeanConsumptions) %>%\n",
    "    pivot_longer(cols=c('MeanImpressions', 'MeanConsumptions'), names_to='variable', values_to='value')\n",
    "\n",
    "data_summary_longer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "fbfdf5c8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA8AAAALQCAIAAADQFY7jAAAACXBIWXMAABJ0AAASdAHeZh94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\nTkSuQmCC",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 360,
       "width": 480
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "options(repr.plot.width = 8, repr.plot.height = 6)\n",
    "\n",
    "# A side by side histogram of consumptions and impressions separated by type\n",
    "data_summary_plot <- data_summary_longer %>%\n",
    "    ggplot(aes(x=Type, y=value, fill=variable)) +\n",
    "    geom_bar(stat='identity', position='dodge') +\n",
    "    labs(x = \"Engagement Type\", y = \"Value\", fill = \"Variable\") +\n",
    "    ggtitle(\"Mean Consumptions and Mean Impressions of Facebook Engagement Types\")\n",
    "data_summary_plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "90e7eb2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "options(repr.plot.width = 8, repr.plot.height = 6)\n",
    "\n",
    "# Mean Consumptions over Mean Impressions ratio plot separated by type\n",
    "data_ratio_plot <- data_summary %>%\n",
    "    ggplot(aes(x=Type, y=MeanRatio, fill=Type)) +\n",
    "    geom_bar(stat='identity', position='dodge') + \n",
    "    labs(x = \"Engagement Type\", y = \"Ratio (Mean Consumptions/Mean Impressions)\", fill = \"Variable\") +\n",
    "    ggtitle(\"Mean Consumtions to Mean Impressions for Facebook Engagement Types\")\n",
    "data_ratio_plot"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf881182",
   "metadata": {},
   "source": [
    "#### 2.4 Plot Observations"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "42d0b8ba",
   "metadata": {},
   "source": [
    "As illustrated above, the impressions far outweigh the consumptions which means on average Facebook displays a post far more often than that post gets clicked. Another key insight is that Videos seem to be shared much more often across Facebook as their mean impressions are by far the highest. However this does not seem to affect that of consumptions, which demonstrates why the Consumption/Impression ratio is more rational than using a single metric.\n",
    "\n",
    "From an eye test, we can **make preliminary assumption that Photo and Status has higher Consumption/Impression ratio than Link and Video type of content.**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4abfe671",
   "metadata": {},
   "source": [
    "#### 2.5 Bootstraping"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "662a2a69-9db1-4e21-a4d0-9e5804dd8ea3",
   "metadata": {},
   "source": [
    "In this section we will calculate 90% confidence intervals for the mean ratio of each type of post. To do this we will sample 1000 times from the original sample using sizes calculated in part 2.3 Data Summary. We also plot the bootstrap distributions to visualize their spread."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "c0d77160-f688-41f7-b316-77d5be7c60b6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Below we take a bootstrap of 1000 resamples for each type of post (Photo, Video, etc.) The sizes were precalculated in the main summary\n",
    "# table and were used again.\n",
    "\n",
    "bootstrap_photo <- data_photo %>% \n",
    "    rep_sample_n(reps = 1000, size = 426, replace = TRUE) %>% \n",
    "    group_by(replicate) %>% \n",
    "    summarize(mean_ratio = mean(ConsumptionsOverImpressions)) %>% \n",
    "    select(mean_ratio)\n",
    "\n",
    "bootstrap_status <- data_status %>% \n",
    "    rep_sample_n(reps = 1000, size = 45, replace = TRUE) %>% \n",
    "    group_by(replicate) %>% \n",
    "    summarize(mean_ratio = mean(ConsumptionsOverImpressions)) %>% \n",
    "    select(mean_ratio)\n",
    "\n",
    "bootstrap_video <- data_video %>% \n",
    "    rep_sample_n(reps = 1000, size = 7, replace = TRUE) %>% \n",
    "    group_by(replicate) %>% \n",
    "    summarize(mean_ratio = mean(ConsumptionsOverImpressions)) %>% \n",
    "    select(mean_ratio)\n",
    "\n",
    "bootstrap_link <- data_link %>% \n",
    "    rep_sample_n(reps = 1000, size = 22, replace = TRUE) %>% \n",
    "    group_by(replicate) %>% \n",
    "    summarize(mean_ratio = mean(ConsumptionsOverImpressions)) %>% \n",
    "    select(mean_ratio)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "5d2785b0",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Warning message:\n",
      "“Removed 2 rows containing non-finite values (stat_bin).”\n",
      "Warning message:\n",
      "“Removed 2 rows containing missing values (geom_bar).”\n",
      "Warning message:\n",
      "“Removed 2 rows containing missing values (geom_bar).”\n",
      "Warning message:\n",
      "“Removed 2 rows containing missing values (geom_bar).”\n",
      "Warning message:\n",
      "“Removed 2 rows containing missing values (geom_bar).”\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABwgAAAHgCAMAAACGislhAAAC91BMVEUAAAABAQECAgIDAwME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\nl7TWmHsWAAAAAElFTkSuQmCC",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 240,
       "width": 900
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "options(repr.plot.width = 15, repr.plot.height = 4)\n",
    "\n",
    "# in this cell we will plot all bootstrap distributions to get an idea of the originial distributions\n",
    "bootstrap_photo_plot <- bootstrap_photo %>%\n",
    "    ggplot(aes(x = mean_ratio)) +\n",
    "    geom_histogram(binwidth = 0.001) +\n",
    "    xlab(\"Mean Consumptions / Impressions\")\n",
    "\n",
    "bootstrap_status_plot <- bootstrap_status %>%\n",
    "    ggplot(aes(x = mean_ratio)) +\n",
    "    geom_histogram(binwidth = 0.001) +\n",
    "    xlab(\"Mean Consumptions / Impressions\")\n",
    "\n",
    "bootstrap_video_plot <- bootstrap_video %>%\n",
    "    ggplot(aes(x = mean_ratio)) +\n",
    "    geom_histogram(binwidth = 0.001) +\n",
    "    xlab(\"Mean Consumptions / Impressions\")\n",
    "\n",
    "bootstrap_link_plot <- bootstrap_link %>%\n",
    "    ggplot(aes(x = mean_ratio)) +\n",
    "    geom_histogram(binwidth = 0.001) +\n",
    "    xlab(\"Mean Consumptions / Impressions\")\n",
    "\n",
    "# Plot the bootstrap distributions side by side\n",
    "plot_grid(bootstrap_photo_plot +\n",
    "              labs(title = \"Bootstrap Distribution for Photo Posts\") +\n",
    "              scale_x_continuous(limits = c(0.1,0.16)),\n",
    "          bootstrap_status_plot +\n",
    "              labs(title = \"Bootstrap Distribution for Status Posts\") +\n",
    "              scale_x_continuous(limits = c(0.08,0.14)),\n",
    "          bootstrap_video_plot +\n",
    "              labs(title = \"Bootstrap Distribution for Video Posts\") +\n",
    "              scale_x_continuous(limits = c(0,0.15)),\n",
    "          bootstrap_link_plot +\n",
    "              labs(title = \"Bootstrap Distribution for Link Posts\") +\n",
    "              scale_x_continuous(limits = c(0,0.06)),\n",
    "          ncol = 4)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cae6b830-075b-4e92-ac40-0151ad2fc262",
   "metadata": {},
   "source": [
    "#### 2.6 Calculating Uncertainty"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "66c443d6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"dataframe\">\n",
       "<caption>A tibble: 4 × 4</caption>\n",
       "<thead>\n",
       "\t<tr><th scope=col>Type</th><th scope=col>ci_lower</th><th scope=col>ci_upper</th><th scope=col>MeanRatio</th></tr>\n",
       "\t<tr><th scope=col>&lt;chr&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th><th scope=col>&lt;dbl&gt;</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "\t<tr><td>Link  </td><td>0.01751523</td><td>0.03580021</td><td>0.02594883</td></tr>\n",
       "\t<tr><td>Photo </td><td>0.10876724</td><td>0.13981007</td><td>0.12364624</td></tr>\n",
       "\t<tr><td>Status</td><td>0.09639651</td><td>0.12061863</td><td>0.10865636</td></tr>\n",
       "\t<tr><td>Video </td><td>0.01697675</td><td>0.08688191</td><td>0.04637671</td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "A tibble: 4 × 4\n",
       "\\begin{tabular}{llll}\n",
       " Type & ci\\_lower & ci\\_upper & MeanRatio\\\\\n",
       " <chr> & <dbl> & <dbl> & <dbl>\\\\\n",
       "\\hline\n",
       "\t Link   & 0.01751523 & 0.03580021 & 0.02594883\\\\\n",
       "\t Photo  & 0.10876724 & 0.13981007 & 0.12364624\\\\\n",
       "\t Status & 0.09639651 & 0.12061863 & 0.10865636\\\\\n",
       "\t Video  & 0.01697675 & 0.08688191 & 0.04637671\\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "A tibble: 4 × 4\n",
       "\n",
       "| Type &lt;chr&gt; | ci_lower &lt;dbl&gt; | ci_upper &lt;dbl&gt; | MeanRatio &lt;dbl&gt; |\n",
       "|---|---|---|---|\n",
       "| Link   | 0.01751523 | 0.03580021 | 0.02594883 |\n",
       "| Photo  | 0.10876724 | 0.13981007 | 0.12364624 |\n",
       "| Status | 0.09639651 | 0.12061863 | 0.10865636 |\n",
       "| Video  | 0.01697675 | 0.08688191 | 0.04637671 |\n",
       "\n"
      ],
      "text/plain": [
       "  Type   ci_lower   ci_upper   MeanRatio \n",
       "1 Link   0.01751523 0.03580021 0.02594883\n",
       "2 Photo  0.10876724 0.13981007 0.12364624\n",
       "3 Status 0.09639651 0.12061863 0.10865636\n",
       "4 Video  0.01697675 0.08688191 0.04637671"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# In this cell we calculate the 90% confidence interval for the original means using the bootstraps calculated previously\n",
    "\n",
    "# Combine all bootstraps to one DF and calculate confidence intervals\n",
    "Confidence_Interval <- bind_rows(mutate(bootstrap_photo, Type = \"Photo\"),\n",
    "    mutate(bootstrap_status, Type = \"Status\"),\n",
    "    mutate(bootstrap_video, Type = \"Video\"),\n",
    "    mutate(bootstrap_link, Type = \"Link\")) %>%\n",
    "    group_by(Type) %>%\n",
    "    summarise(ci_lower = quantile(mean_ratio, 0.05),\n",
    "              ci_upper = quantile(mean_ratio, 0.95)) %>%\n",
    "    bind_cols(select(data_summary, MeanRatio))\n",
    "\n",
    "Confidence_Interval\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d661bd88",
   "metadata": {},
   "source": [
    "**Analysis**\n",
    "\n",
    "- Unsurprisingly the **confidence interval for Video is the largest**. This is because we have only 7 observations with the type video and as a result we need a wider confidence interval. \n",
    "- Surprisingly **Link has the smallest confidence interval even though we have the most observations of type Photo**. This could be a result of random chance in the bootstrapping samples.\n",
    "- Since the **confidence interval for Photo and Status is clearly above the range of confidence interval for Link and Video**, we now have more conviction to say that these two are the more effective content types"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2805004e",
   "metadata": {},
   "source": [
    "## 3. Methods: Plan"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1029910e",
   "metadata": {},
   "source": [
    "#### 3.1 Methods & Plan"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01a0d14c-01fb-4350-8798-8ae6acb96fa8",
   "metadata": {},
   "source": [
    "Our project aims to tackle a critical issue in digital advertising, and we aim to provide a more in-depth perspective built on research paper by Moro (Moro et al., 2016). We have made taken steps to select a representative data set, clean & wrangle data and identify the most relevant variable to analyze. Through basic data exploratory and bootstrapping, the preliminary assumption have aligned with the statistical evidence. \n",
    "  \n",
    "  However, since there are variance between our bootstrap estimates and population parameter, our charts and estimates will not suffice. Therefore, we plan to communicate the uncertainty of our estimates by constructing and visualizing 95% confidence intervals for each content type. We then will proceed with hypothesis testing to get the statistical evidence to confirm that Photo and Status are the most effective content types."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c814d8c-987f-4929-8dfc-58a955ccaec8",
   "metadata": {},
   "source": [
    "#### 3.2 Implication of Research"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "207b73db-7fc4-4580-8207-55885db1b1d5",
   "metadata": {},
   "source": [
    " Understanding which type of content will engage customers the most will allow companies to best allocate their advertising budget. There are two future questions to analyze more:\n",
    "   - Would different types of products being advertise (such as clothes, food, education) have an effect on Consumption/Impression ratio?\n",
    "   - Would video and photo still be the most effective type of content on other platforms? For example, we have Instagram story, TikTok short videos and even LinkedIn advertisement. All can change the dynamics of a social media platform and thus affect what type would work best."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4503dda",
   "metadata": {},
   "source": [
    "## 4. References"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32176162",
   "metadata": {},
   "source": [
    "1. Moro, S., Rita, P., & Vala, B. (2016). Predicting social media performance metrics and evaluation of the impact on brand building: A data mining approach. *Journal of Business Research*, 69(9), 3341-3351. https://doi.org/10.1016/j.jbusres.2016.02.010 \n",
    "2. Shaw, A. (2018). How social media can move your business forward. *Forbes*. https://www.forbes.com/sites/forbescommunicationscouncil/2018/05/11/how-social-media-can-move-your-business-forward/?sh=f3f47264cf2b\n",
    "3. Sheerin, ., Patel, S., Oundjian, A. & Leon, H. (2020). Digital marketers race to meet the needs of a new era. *Boston Consulting Group*. https://www.bcg.com/publications/2020/digital-marketers-race-to-meet-the-needs-of-growing-online-agendas\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "R",
   "language": "R",
   "name": "ir"
  },
  "language_info": {
   "codemirror_mode": "r",
   "file_extension": ".r",
   "mimetype": "text/x-r-source",
   "name": "R",
   "pygments_lexer": "r",
   "version": "4.1.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}