Professor Andrew Garrard

Professor Andrew Garrard

I'm the Head of Department in the University of Sheffield's £81 million teaching facility, the Diamond. I'm an engineer that teaches fluids, thermo and CFD. Follow me on twitter @fluidsandthermo

Location United Kingdom


  • It is an easy thing to scan your report and check for errors. The content of reports is usually so much harder to produce than the report itself that it is worth taking a bit of extra time to ensure it isn't ruined by poor writing and presentation technique.

  • Hopefully following the course will help with that. One of my objectives is to not simply tell people how to do things, but explain why it is a good idea to do things. If you feel any parts of the course are lacking in this regard, please do let me know.

  • Thanks so much everyone.

  • Here is a general rule (which is a way of saying one that doesn't always apply): Graphs show trends in numbers, tables show the actual numbers.

  • I hope we can all agree - this is a very ugly graph!

  • This is my favourite answer so far. If you repeat an experiment with a different "controlled" variable (in this case, position) then this can be represented as a different series on the same set of axes. It is very important that the reader can tell one series from another with, say, colours, shapes, dots...etc and there is an appropriate key.

  • I'm going to suggest that that the error analysis needs to be done before the number of SF is quoted in the report. Put simply, the SF implies the precision of the number quoted. Without knowing the uncertainty of the measurement, you can't determine the error and can't accurately specify the SF.

  • Data should be recorded in the units of measurement, but this isn't necessarily the best way to present it in a report.

  • The key is to think of your reader and be concise. I'd err on the side of less rather than more in the results section and, as you say, put supplementary information in the appendix for completeness.

  • Should the raw data from the experiment be recorded, or the processed data?

  • Well put

  • I think the key words you use here are "possible sources". If there is conjecture about why a result is they way it is, this belongs in the discussion, not the results, as it isn't hard fact.

  • Would the procedure not do that?

  • You are correct, of course, that results could be interspersed with other sections. This is sometimes done in very short reports or if the material demands it. But generally, yes, the results presented without clutter can increase understanding but also allows people to skip directly this section if that is all they are interested in.

  • "without interpretation" - this is very important. There is a place to interpret, and that is the discussion. The results section should only contain facts.

  • We also have that expression in the UK.

  • This is a great point. If it is faster for the reader to obtain information from a figure rather than text, then include the figure.

  • I agree, these isn't a correct amount. I feel the trick is to justify EVERY figure, and if you can't justify including it, don't!

  • Do feel free to comment and provide feedback on one another's work. It will help everyone to learn.

  • Thanks for the +ve feedback

  • My feelings about the importance of the procedure section are related to the results. The way a study is conducted can influence the results obtained, so it is critical for the reader to know those methods in order to critically evaluate the results.

  • Agreed. The point I was trying to make is this section is focusing on the aspects of a report that are "around" the figure, rather than the figure itself. We have another section for "in the figure".

  • Yeah, but the captioning is good!

  • When I was builidng this course, I wouldn't have necessairly had pictures at the start of each of these web pages. They don't add any technical detail and their function is to increse readability. While I wouldn't put a picture in a technical report just for the sake of it, there could be reasons beyond just information transmittion.

  • Totally agree.

  • I think this is an excellent answer.

  • Thanks for all of these, everyone. They are all very good. One good attribute of Declan's answer was to include the values of the settings. When the plan of what the procedure would be is given, the settings wouldn't have been known as the experiment wasn't done. To write up the procedure, it is a record of what happened, so the specifics can be included.

  • The amount that is "too much" or "too little" is a matter of judgement - and you can improve with more practice.

  • This is so important. I think it is covered in week 6. I often type out lots of text, read it myself and don't spot any mistakes, and then others read it and find things that are obvious errors. When you read your own work, you tend to read the words you think you typed rather than those you did.

  • Great summary. I like to use the word succinct.

  • There is something called a "personal pronoun". These are words like "I", "we", "you"...etc. In scientific terms, the operator is usually not that import, if you read volt meter it is no different to me reading a volt meter, so the sentence should say "the voltmeter was read". When I'm marking student's work, I quickly scan for personal pronouns in the text...

  • This is a good point.

  • This is correct. I'm not a big advocate of following conventions for no reason. If you don't understand why conventions are there, it is a good idea to follow them because they are probably there for a good reason. But if you do know why they are used, but in your case it isn't appropriate or detrimental to follow them - then don't.

  • How do you know they are not influenced by the manufacturer?

  • I'm scientific research you are allowed to offer opinion. One of the jobs of a lit review is to establish the range of opinion at the state of the art.

  • You raise an interesting point. The words for things are sometimes not that important as what they represent. If you understand the distinction and feel that knowing the difference is of value when thinking about how you articulate to your reader, who cares what they are called or if you use those words in your reports.

  • I really like the analogy in point 1

  • Thanks for all the comments everyone. I'm open to helpful suggestions to changes though, so don't hold back. Maybe more will come in future weeks!

  • There is no right or wrong answer, but you should think whenvever you consider including anything. If you feel your reader would benifit from the background theory you are employing, or you are demonstrating your understanding, then it should be included. If it is of little value to the reader, in your opinion, then don't. If you are unsure, I'd suggest...

  • I think sometimes it isn't obvious. For example, if you are using something so well know, such as Newton's second law F=ma, do you need to cite the source?

  • very deep!

  • Well - now you can!

  • Yep, referencing tools are very useful for managing the process. The tool can't help you establish when to reference or you would want to, though!