Preparing to take the SAT in the coming months? You’ll want to complete full practice tests as part of your test prep process—preferably ones as similar to the real SAT as possible. If you’ve been scouring the internet for practice test material, you may have read that Google Gemini and the Princeton Review recently teamed up to launch full-length SAT practice tests. We wanted to see if the tests were worthy of all the buzz, so our test prep experts took a deep dive into them. Here’s what we found.
The Testing Experience
Gemini’s SAT practice tests deviate from the real SAT in some significant ways. The Gemini platform has a very different look and feel from what you’ll see on the official College Board Bluebook app:
- The usual platform tools are missing. These exams do not include the highlighter, answer eliminator, and Desmos calculator provided on the real test.
- You can choose to see the correct choice immediately after selecting an answer. You can select this option—which is not offered on official College Board practice exams—at the start of a test. On multiple-choice questions, the correct answer appears immediately after you click on an answer choice, even if you select the answer accidentally, so you cannot change your answers or go back to correct previous work. Student-produced response questions do include a “check answer” button, however.
- Scroll bars are difficult to locate. When you need to scroll down on a page, you may need to click around on the page in order to make the scroll bar appear.
- If you get stuck, you can choose to see hints. This is another option that you can select at the start of each test, and that is not offered on College Board practice tests. If you select this option, you’ll see a small “show hint” prompt below each question.
- Answer explanations are not easy to follow. The explanations are not written to help you learn each concept, but rather are presented as a bulleted list of sometimes confusing steps to solve or answer the questions. Question types are also often incorrectly categorized.
- Significant technical glitches can occur. One tester was only two questions away from the end of the exam when they were automatically signed out of Gemini. Upon signing back in, the tester was prompted to begin an entirely new SAT. According to Gemini, the test modules are session-bound, so once the link expires or the connection resets, the system defaults to a new exam.
Reading & Writing
We discovered a few Reading & Writing section-specific issues on Gemini’s SAT practice tests that would seriously impact the student experience.
- Question types do not appear in the same order that they do on the SAT. Real SAT Reading & Writing sections include vocabulary-in-context, reading comprehension, grammar, transitions, and rhetorical synthesis questions, in that order. On these exams, however, the question types appear randomly, and many testers saw reading comprehension questions at the start of the section.
- Some question types may not appear at all. One tester, for example, did not see a single rhetorical synthesis question in an entire module.
- The balance of question types is not the same as it is on the SAT. Some testers saw far more vocabulary questions and far fewer grammar questions than normal. Others saw far more reading comprehension questions than normal. Because some question types take longer to complete than others, this can seriously impact your pacing.
- Exams include grammar concepts not tested on the SAT. Testers reported seeing questions about article use and a non-standard grammar question testing transitions with a “none of the above” option—question types you will not see on the SAT.
- The exams feature more literary texts (literary prose, poetry, and plays) than usual. These tests included more than double the number of literary texts that appear on the real SAT.
Math
The Math sections also presented our testers with a number of unfortunate issues.
- Math sections tested concepts that do not appear on the SAT. These included imaginary numbers, the law of cosines, and logarithms.
- Some test questions contained typos. In some cases, those typos made questions impossible to solve.
- Poor question construction and irrelevant information made questions unnecessarily difficult. Imprecise wording made certain questions unsolvable. Other questions included completely unnecessary givens, followed by a straightforward question prompt, which is out of step with what you’ll see on the real SAT. For example, in one case, specific values of natural logarithms were given, even though they were immaterial to solving the question (and the real SAT does not test logarithms anyway).
- The order of difficulty varies from what you’ll see on the SAT. On the real SAT, you’ll receive a first Math module with a mix of easy, medium, and difficult questions. If you score well on that first module, you’ll receive a harder second module, and if you struggle with the first module, you’ll receive an easier second module. Even though each of our testers answered enough questions correctly to receive a difficult second module, we found that module two was significantly easier than module one.
The Takeaway
AI has stepped up its game in the past few years, and chances are it’ll keep improving. However, are AI-generated practice tests ready to become your go-to for SAT prep? No. The SAT is a high-stakes test, and you want to go into your test confident that you know the exam inside and out and won’t encounter any surprises on test day. To do that, you’ll want to prep with materials as close to the real thing as possible.
Need help prepping for your next SAT? Private Prep offers a full line-up of practice SATs that accurately mirror what you’ll see on test day, and our experts are here to help. Contact us today.
