hankandteri

On the Silhouette now in the middle of the Atlantic. We upgraded to premium Wi-Fi during a flash sale, and so glad we did. Folks on board with basic have told us they aren’t happy with it. We can post and watch videos even out here. It’s really amazing.

We missed three ports on our early summer cruise on Oceania. Like you, we were repeaters on O (prior sailing was prepandemic). Long story, but there was no real reason for two of those three misses (Lisbon and Porto). I’ve noticed that NCL-owned ships seem to skip a lot more ports than other lines. I thought the food on O was good to great, but I won’t be going back because of the port skipping shenanigans. I’m not that picky, but I expect to be taken where you promised to take me, and I expect to be fed reasonably well along the way. Why is that so difficult?

I tried to like your post, but got an automated message that I couldn’t interact with it. I couldn’t find an objectionable word phrase or thought in it.

Wow! Just wow! MSC overall (entire fleet, not just Divina) is already quite a bit lower than the other mass market lines, so for one of their ships to drop an entire point in a year is quite the accomplishment. I keep reading about how nice the YC is and how beautiful MSC ships are, but also how crowded they are and how the food and service aboard can be less than stellar, to put it kindly. I’m a self described numbers nerd but I’m the kind who likes to “find out for myself” and test the numbers and comments against reality. MSC is the one line I haven’t sailed and wouldn’t consider right now based strictly on reviewer comments and scores.

This thread kind of took on a life of its own, so I can’t blame you for not wanting to wade through all all of it. After others asked, I also did the same calculation for three other lines (Carnival, HAL and Virgin) to see if that recent dip in CC reviewer rating was impacting all cruise lines. Carnival recent review scores also dropped from their historic average, nearly as much as Celebrity’s did. HAL reviews were essentially flat. Virgin’s recent reviewer scores were up significantly over their already high benchmark rating.

It’s time to put the old supply chains excuse to bed. If you’ve visited a grocery store anywhere in Europe or the US lately, you know that produce and quality meats are readily available. Other lines (see Carnival post above) are having no difficulty sourcing quality foodstuffs.

With the mentions of Gary Bembridge, Virgin, and the usefulness of Cruise Critic reviews on this thread, here’s something that combines all three. Mr. Bembridge just dropped a podcast episode that explains why Virgin, Disney, and Viking are rated so highly on Cruise Critic. He goes on to give some very interesting analysis of Celebrity at the end. His analysis is insightful but doesn’t take into account recent operational changes. Worth a listen. https://sites.libsyn.com/49893/the-cruise-lines-everyone-is-raving-about-right-now-and-why-podcast-382

The NPS is likely derived from a question embedded in the post cruise survey. That’s how I’d do it if I were them.

Which raises the next important statistical topic—confirmation bias—something you see a lot of on all CC boards.

Love it. Did you know that 73% of statistics are made up on the spot?

And we will be aboard Silhouette next week. Like you, I’m hoping to report that we had a wonderful experience when we return, and that the glitches you experienced in June have been addressed. Our cruise is also sold out, so the crew will be pushed. Even if things aren’t perfect (and let’s face it, no cruise is flawless in the best of times), I’m positive we will enjoy it and meet a lot of wonderful people on board.

If someone wants to send me the NPS raw data, I’d be happy to take a look at it, but that isn’t going to happen. My gut instinct is that it that it would tell the same tale, but no company would ever share precious proprietary data like that, especially if the story it told was bad.

Much better!

Let’s pretend for a minute that our statistics/market research degree awarded by Google really gave this horrifying answer. The thing is my sample size was 100% of the population I was measuring, far above the bizarre 5 to 30% minimum stated. My universe was people who submitted X reviews on CC for all of Sept. and the last part of Aug., and I measured every bit of it. Then I measured that perfect sample against the average of 100% of those who have ever submitted X reviews on CC. The sample you’re demanding isn’t available because your universe could only exist if 100% of X cruisers submitted reviews on CC. You can only measure a population that exists. You could probably Google that. So, while my little exercise may not accurately reflect the hypothetical “normal” X cruiser (which was probably your main point until you went down the statistical validity rabbit hole), it perfectly captures the complete cohort I was measuring with 100% accuracy. You are saying my results aren’t statistically valid until a sufficiently large sample comes from a population that could never exist. I measured the population created by parameters I chose, and I measured every bit of it. You can argue about whether the measurement means anything or the size of the population measured. Fine, but statistical validity is not an issue.

I feel the same way when it comes to toasters, dentists, and restaurants, but if you need to see a 4.5 (or even a 4.0) rating for a cruise line on this site in order to book, your options will be limited right now. For every cruise line I looked at, at least a few people were assigning 1s in the most recent reviews. Scores like that strain credulity even in the current environment. A single 1 requires eight 5s to get an average above 4.5 with no other ratings under 5. That’s a daunting task even with a line that’s totally on its game.

You lost me at "I'm not a numbers guy."

Statistics say nothing of the sort. In fact, statistics don't say anything at all, but statisticians do. Your source please, sir or madam. The size of your universe isn't all that relevant to the sample needed to gain insight. Plenty of clinical drug trials involve very small samples, and loads of valid market research efforts are done with relatively tiny sample sizes, depending on the objective. I've conducted a lot of that research myself over several decades. I don't pretend my little exercise is scientific, only that it appears to indicate something. Believe that there's meaning in the numbers or validity to my methodology or don't. It's all good.

Exactly! Peaches knows her stuff. What I shared is more than anecdotal and less than scientific. It was mostly a quick reality check I did for my own benefit born out of curiosity. The result surprised me so much that I decided to share them here. Sixty is a pretty healthy sample for something like this, though. I could have gone back a little further to capture all of August in addition to September, but the totals I was getting were pretty consistent from screen to screen. You might not like the numbers (I certainly don't) and you might be able to explain them away or rationalize them, but they are what they are. That's the cool thing about data. It doesn't have an agenda. The agenda comes in when you begin to manipulate or make more of it than is there. I didn't do anything tricky or that anyone else here couldn't reproduce on their own in a matter of minutes. The numbers speak for themselves. I leave it to others to decide what, if anything, they mean. Clearly, some of you think they align with reality and others of you don't. It's all fine by me.

Totally agree. It would be very interesting to see how any cruise line's score averages at any given point moved when layered against the timing of operational changes (such as improvements or declines in food or entertainment).

If you read the entire thread, you'll see that several people were speculating that all major cruise lines would have lower recent ratings. It was a fair point, and I wanted to know if that was true or not, so I checked the data for a few others (Carnival, HAL, Virgin) and shared the numbers here. Carnival was also way down, HAL was virtually flat, Virgin was significantly up.

Yes!!!!! You totally get it. This was simply a quick and dirty exercise to see if anything popped out, and it did. I'll leave it to others to determine the reasons and meaning (if any). If anyone really cared, they could track these results over time and also check against potential variables like seasonality of reviews (is the average review higher in some seasons across all years?) and all kinds of other things. I didn't do any of that; but if I were in the executive suites in Miami, I would assign a bright intern to the task of getting all this data for X and their major competitors and start building out a spreadsheet with some serious crosstabs built in.

Virgin Voyages has a historical average of 4.1 and their most recent 60 reviews averaged 4.5, so a significant uptick in recent passenger satisfaction. Not every cruise line is disappointing their customers right now.

I just did Carnival (3.8 to 3.0) and HAL (3.9 to 3.8) Any other nerd here can do what I did for any cruise line. It only takes a couple of minutes. Cruise Critic loads 10 reviews on each screen so all you need to do is add the total number of dots for the first six screens and divide by 60 to get the current score I was measuring.

Again, I admitted that I didn't compare current vs. historical averages for any other line in my initial report. I was only concerned about recent Celebrity reviews. Just for fun, I just did the same exercise for Carnival and HAL (last 60 reviews vs. historical average) Carnival also dropped and by way more than I was expecting. They went from 3.8 to 3.0--almost as much as X dropped. HAL, on the other hand, went from 3.9 to 3.8, which isn't much of a change at all.

Comparing all reviews to recent reviews is apples to apples in that regard. You'd have axe grinders and glowers in both cohorts.