Analysis of ‘Popular Website Traffic Over Time ’ provided by Analyst-2 (analyst-2.ai), based on source dataset retrieved from https://www.kaggle.com/yamqwe/popular-website-traffice on 13 February 2022.

--- Dataset description provided by original source is as follows ---

About this dataset

Background

Have you every been in a conversation and the question comes up, who uses Bing? This question comes up occasionally because people wonder if these sites have any views. For this research study, we are going to be exploring popular website traffic for many popular websites.

Methodology

The data collected originates from SimilarWeb.com.

Source

For the analysis and study, go to The Concept Center

This dataset was created by Chase Willden and contains around 0 samples along with 1/1/2017, Social Media, technical information and other features such as: - 12/1/2016 - 3/1/2017 - and more.

How to use this dataset

• Analyze 11/1/2016 in relation to 2/1/2017
• Study the influence of 4/1/2017 on 1/1/2017
• More datasets

Acknowledgements

If you use this dataset in your research, please credit Chase Willden

Start A New Notebook!

--- Original source retains full ownership of the source dataset ---

Comparison of user, site, and network-centric approaches to web analytics data collection showing advantages, disadvantages, and examples of each approach at the time of the study.

Notice: You can check the new version 0.9.6 at the official page of Information Management Lab and at the Google Data Studio as well.

Description of the Report and Topic Justification

Now that the ICTs have matured, Information Organizations such as Libraries, Archives and Museums, also known as LAMs, proceed into the utilization of web technologies that are capable to expand the visibility and findability of their content. Within the current flourishing era of the semantic web, LAMs have voluminous amounts of web-based collections that are presented and digitally preserved through their websites. However, prior efforts indicate that LAMs suffer from fragmentation regarding the determination of well-informed strategies for improving the visibility and findability of their content on the Web (Vállez and Ventura, 2020; Krstić and Masliković, 2019; Voorbij, 2010). Several reasons related to this drawback. As such, administrators’ lack of data analytics competency in extracting and utilizing technical and behavioral datasets for improving visibility and awareness from analytics platforms; the difficulties in understanding web metrics that integrated into performance measurement systems; and hence the reduced capabilities in defining key performance indicators for greater usability, visibility, and awareness.

In this enriched and updated technical report, the authors proceed into an examination of 504 unique websites of Libraries, Archives and Museums from all over the world. It is noted that the current report has been expanded by up to 14,81% of the prior one Version 0.9.5 of 439 domains examinations. The report aims to visualize the performance of the websites in terms of technical aspects such as their adequacy to metadata description of their content and collections, their loading speed, and security. This constitutes an important stepping-stone for optimization, as the higher the alignment with the technical compliencies, the greater the users’ behavior and usability within the examined websites, and thus their findability and visibility level in search engines (Drivas et al. 2020; Mavridis and Symeonidis 2015; Agarwal et al. 2012).

One step further, within this version, we include behavioral analytics about users engagement with the content of the LAMs websites. More specifically, web analytics metrics are included such as Visit Duration, Pages per Visit, and Bounce Rates for 121 domains. We also include web analytics regarding the channels that these websites acquire their users, such as Direct traffic, Search Engines, Referral, Social Media, Email, and Display Advertising. SimilarWeb API was used to gather web data about the involved metrics.

In the first pages of this report, general information is presented regarding the names of the examined organizations. This also includes their type, their geographical location, information about the adopted Content Management Systems (CMSs), and web server software types of integration per website. Furthermore, several other data are visualized related to the size of the examined Information Organizations in terms of the number of unique webpages within a website, the number of images, internal and external links and so on.

Moreover, as a team, we proceed into the development of several factors that are capable to quantify the performance of websites. Reliability analysis takes place for measuring the internal consistency and discriminant validity of the proposed factors and their included variables. For testing the reliability, cohesion, and consistency of the included metrics, Cronbach’s Alpha (a), McDonald’s ω and Guttman λ-2 and λ-6 are used.
- For Cronbach’s, a range of .550 up to .750 indicates an acceptable level of reliability and .800 or higher a very good level (Ursachi, Horodnic, and Zait, 2015). - McDonald’s ω indicator has the advantage to measure the strength of the association between the proposed variables. More specifically, the closer to .999 the higher the strength association between the variables and vice versa (Şimşek and Noyan, 2013). - Gutman’s λ-2 and λ-6 work verifiably to Cronbach’s a as they estimate the trustworthiness of variance of the gathered web analytics metrics. Low values less than .450 indicate high bias among the harvested web metrics, while values higher than .600 and above increase the trustworthiness of the sample (Callender and Osburn, 1979). -Kaiser–Meyer–Olkin (KMO) and Bartlett’s Test of Sphericity indicators are used for measuring the cohesion of the involved metrics. KMO and Bartlett’s test indicates that the closer the value is to .999 amongst the involved items, the higher the cohesion and consistency of them for potential categorization (Dziuban and S...