Overview of Statistical Resources
Common sources of published statistics that are accurate & meaningful are:
- Government agencies (Countries, States, Counties and Cities).
- National & international based Trade Associations & Professional Societies.
- Think Tanks and Lobbying Organizations. These groups tend to have strong underlying biases within the organizations themselves, which may not be evident from the data that they offer.
It is highly unlikely that private for profit corporations will share much of their internally generated statistics with the public at large, for free. Instead, for profit corporations are more likely to only release carefully selected excerpts of their internal data, and in very controlled situations. Corporations are not going to give away free access to their crown jewels!
Typically, government agency statistics are more likely to be freely available to anyone. While statistics that were compiled by private organizations are more likely to have price tags attached to them. One big exception to this are "publicly traded corporations" which are under a combination of mandates by government agencies & stock exchanges, to disseminate specific types of business information to their existing & potential shareholders. This type of corporate information is typically found in Annual Reports to Shareholders, and also in SEC Forms #10-K. This type of information is covered in more detail in a LibGuide on Annual Reports to Shareholders and 10-K Reports.
If a for-profit company or industry association is offering large datasets separate from "shareholder reports" or "SEC reports", then you should be suspicious that they might have some biases built into them.
Government statistics tend to focus on who currently is using a specific governmental agency or service, and have low likelihoods of containing forecasts of future usage/demand. Government based statistics are also more likely to be looking at broad macro-economic indicators.
On the other hand, trade associations; professional societies; lobbying organizations and private think tanks are more likely to commission surveys about prospective use of new technologies, emerging services, or try to anticipate potential changes in usage/demand. These private sources of statistics are more likely to contain forecasts of future usage/demand. But these intermediate level organizations will not distribute statistical reports if the findings may lead to a degraded image or profits of their underlying professional members or corporate supporters.
Technically speaking, no author or publisher can be totally immune to bias. Most arms of the U.S. Government's agencies are considered to be relatively low on bias, and rated relatively high in terms of accuracy & reliability of their statistical information.
Therefore you should judge each statistical report, each author, and each publisher on its own merits, before jumping to any conclusions about levels of potential or inherent bias.
As you use these statistical resources, please be warned that many of the underlying statistics do not show any author's names. Other times, the document titles & authors are listed at the bottom of a chart, rather than at the top of the page. You need to be very careful in understanding the proper way to cite these underlying statistics & documents.