class: center, middle, inverse, title-slide # Writing Up Results ### Jennifer Mankin ### 18 - 19 March 2021 --- <script type="text/javascript"> setup() </script> ## Today's Tasks - What Goes in Results? - Have a Look at the Red and Green Results - Writing Up --- ## Timeline of the Report - Week 6: Green study analysis - Last week: Red study analysis - This week: Writing up Results - Week 9: Submit! --- ## What Goes in Results? - A clear, narrative explanation of all of your analyses, and what you found - The results of the statistical test you performed - An interpretation of the results of this test in simple language - A figure of the data - How to report [the Green Study](https://and.netlify.app/tutorials/06/#reporting-the-results) and [the Red Study](https://and.netlify.app/tutorials/07/#reporting-the-results) <br><br> .center[Full information on the [Lab Report Information and Resources page on Canvas](https://canvas.sussex.ac.uk/courses/12727/pages/lab-report-information-and-resources)] --- ## What Goes in Results? - Captured primarily by the **Analysis** element of [the marking criteria](https://canvas.sussex.ac.uk/courses/12727/pages/marking-and-feedback-information) - First grade band criteria: > Excellent data analysis and presentation evidenced by a correctly-executed analysis that is appropriate to address the research question, and by a coherent presentation that implies a full understanding of visualizing, analysing, describing and interpreting data. -- - Correct and coherent analysis for the research question - What <font color="maroon">analysis</font> was performed and what the results were: significance tests, CIs - <font color="orange">Describing</font> what you found: means and SDs or counts - <font color="teal">Visualising</font> the data: the figure - <font color="purple">Interpreting</font> the data: what this pattern of results means --- ## Have a Look! - Open the [Red Study paper](http://www.baomee.info/pdf/Romance/0.pdf) and [Green study paper](http://people.bu.edu/msoren/Griskevicius.pdf) - Read Experiment 1, Results and Discussion, in both - Can you find the <font color="maroon">analysis</font>, <font color="orange">descriptives</font>, <font color="teal">visualisation</font>, and <font color="purple">interpretation</font> in each? - Take notes and work together! --- ## Have a Look: Red "An independent-samples *t*-test examining the influence of color condition on perceived attractiveness revealed a significant color effect, *t*(20) = 2.18, p < .05,d = 0.95(see Figure 1B). Participants in the red condition, compared with those in the white condition, rated the target man as more attractive (*M* = 6.79, *SD* = 1.00, and *M* = 5.67, *SD* = 1.34, respectively). None of the participants correctly guessed the purpose of the experiment. Thus, the results from this experiment supported our hypothesis and suggested that color influences participants’ ratings without their awareness." -- .center[<font color="maroon">Analysis</font> | <font color="orange">Descriptives</font> | <font color="teal">Visualisation</font> | <font color="purple">Interpretation</font>] <font color="maroon">An independent-samples t-test examining the influence of color condition on perceived attractiveness revealed a significant color effect, t(20) = 2.18, p < .05,d = 0.95</font> <font color="teal">(see Figure 1B)</font>. <font color="orange">Participants in the red condition, compared with those in the white condition, rated the target man as more attractive (M = 6.79, SD = 1.00, and M = 5.67, SD = 1.34, respectively)</font>. None of the participants correctly guessed the purpose of the experiment. <font color="purple">Thus, the results from this experiment supported our hypothesis and suggested that color influences participants’ ratings without their awareness.</font> --- ## Have a Look: Green ...The key prediction in the experiment was that activating status motives should increase the likelihood of choosing the green product relative to the same green product in the control condition. As seen in Figure 1, whereas 37.2% of participants chose the green car in the control condition, 54.5% of participants chose it in the status condition, `\(\chi^2\)` (1, N = 168) = 4.56,p = .033, `\(\phi\)` = .165...In summary, activating status motives led people to increase the likelihood of choosing proenvironmental green products over more luxurious nongreen products. Consistent with predictions, status motives increased people’s tendencies to forgo luxury when given the opportunity to choose an equally priced green product that could signal one’s prosocial nature... -- .center[<font color="maroon">Analysis</font> | <font color="orange">Descriptives</font> | <font color="teal">Visualisation</font> | <font color="purple">Interpretation</font>] ...The key prediction in the experiment was that activating status motives should increase the likelihood of choosing the green product relative to the same green product in the control condition. <font color="teal">As seen in Figure 1</font>, <font color="orange">whereas 37.2% of participants chose the green car in the control condition, 54.5% of participants chose it in the status condition</font>, <font color="maroon"> `\(\chi^2\)` (1, N = 168) = 4.56,p = .033, `\(\phi\)` = .165 </font>...<font color="purple">In summary, activating status motives led people to increase the likelihood of choosing proenvironmental green products over more luxurious nongreen products. Consistent with predictions, status motives increased people’s tendencies to forgo luxury when given the opportunity to choose an equally priced green product that could signal one’s prosocial nature</font>... --- ## Pro-Tips for Great Results - Keep it simple and clear - Be succinct! Most of your word count should be in the Discussion -- - Use standardised reporting formats - Yes: "A correlation analysis indicated that strength of sexual and romantic attraction were significantly positively correlated (*r*(299) = .51, *p* < .001, 95% CI [.42, .59]).” - No: "There was a significant positive correlation with an *r* of .51. There were 299 degrees of freedom. The value of *p*...." -- - Don't explain basic statistical concepts or common analyses - Yes: "A correlation analysis indicated that strength of sexual and romantic attraction were significantly positively correlated (*r*(299) = .51, *p* < .001, 95% CI [.42, .59]).” - No: "Correlation is a standardised measure of how much two variables covary that is significant if the correlation coefficient *r* is unlikely to occur if the null hypothesis of no relationship at all is true..." 😓 😴 --- <br><br><br><br><br> ## Questions? --- ## Support Reminders - Resources - [Lab Report Information and Resources](https://canvas.sussex.ac.uk/courses/12727/pages/lab-report-information-and-resources): complete guide to the report - Practicals and tutorials: Week 6 (Green) and Week 7 (Red) - Academic Advising session on Discussions - Getting Help - [Post on Piazza](https://canvas.sussex.ac.uk/courses/12727/external_tools/6455) - [Come to Lab Report Drop-ins](https://canvas.sussex.ac.uk/courses/12727/pages/r-help-desk-drop-ins) - We will not answer lab report questions via email! --- ## Writing Up - Create everything you need in R - Work through the [Red](https://and.netlify.app/practicals/07/worksheet/) or [Green](https://and.netlify.app/practicals/06/worksheet/?sol=true) worksheet - Compare to the [Lab Report Info page](https://canvas.sussex.ac.uk/courses/12727/pages/lab-report-information-and-resources) - Write a short outline that includes your <font color="maroon">analysis</font>, <font color="orange">descriptives</font>, <font color="teal">visualisation</font>, and <font color="purple">interpretation</font> - Copy over any numbers and figures if necessary - See [this week's seminR](https://and.netlify.app/seminr/) for more help! - Fill out your outline into a clear and succinct summary --- class: last-slide <br><br><br><br><br> # And that's it!