# Week 3 Reproducibility Plan template
# Tutorial Tuesday 4-6 pm Group 4

## Summary and Reactions

# Joli

## Summary
> [name=junekam17] Really clear summary of the study, clearly and concisely explaining each section of the aims, methods, results, and conclusions. It may be worth briefly mentioning the analysis strategies used by the experimenters.

#### Aims/rationale:
Investigating whether belief of previous contract of COVID-19 was positively associated with lower engagement with lockdown measures
> I think it would be better to clarify that this was only a single hypothesis of the experiment. While probably the most critical hypothesis for the implications suggested, I think it is important to clarify the belief of prior COVID infection on self-reported behaviour more generally – rather than specifically 'lower engage with lockdown measures' – so that the findings of this and future research can appear generalisable to behavioural implications holistically of this particular belief. [name=shaily]

#### Methods:
Online survey (which was part of a wider experimental study) asked participants about their behaviour on the past 7 days, to rate worry about COVID-19, rate risk of covid-19 to themselves and others, belief about common symptoms of coronavirus, extent of agreement about having immunity to COVID, about being tested to covid.
> [name=ChrisShen] I think this is a good, concise summary of the method used. Especially covering the questions being answered. I think mentioning the scales they used for answering would be important.
#### Results:
Quarter participants (24.3%) believed they had COVID with only 4% having reported a positive result. Compared to those who did not believe they had contracted COVID-19, those with this belief had a higher report of belief of immunity to COVID, lower report of following lockdown measures, lower worry about COVID-19 and lower knowledge about symptoms of COVID-19.

> [name=junekam17]Did the authors mention any driving factors that may be contributing to this large proportion of those that believed they had had COVID-19? Further, clarifying what was meant by adherence to lockdown measures may be an important point for readers, as well as the wider implications this may have on the general population.

> I also think that if mentioning the result of 'lower knowledge about symptoms of COVID-19', it would be good to clarify which variable/hypothesis this relates to which the experimenters were measuring as it could be confusing for a reader where this hasn't already been introduced – which was to assess the likelihood of misdiagnosis. This was not necessarily conducive to their hypotheses, but more so as an extra variable to interrogate the construct of misdiagnosis so I think it could be confusing groupign it with the other results which are more relevant to their hypotheses. [name=shaily]

#### Conclusions:
These papers results show that people who believe they had COVID are less likely to adhere to protective behaviours, but more likely to believe they have immunity. Even the result of an antigen test was not reflected in people’s belief about whether they have COVID-19. Following from this paper, there is a need for clear communication to this group of people about the importance of engaging in safer measures.
> [name=ChrisShen] I think this paper did a good job reporting the limitations that it had. Some of the could have had significant effects on the overall results such as the sample being surveyed and potential self-report bias.

## Reaction 1
This paper reminded me of the Health Beliefs Model. Decreased worry about COVID-19 correlated with thinking you had COVID-19, so perceived threat about COVID leads to behaviour makes sense. However, little evidence for association between belief about COVID and perceived risk - paper argues about this not making sense which I agree with, since there should also be less perceived risk. Whilst the main finding about “adherence to lockdown measures for COVID-19 is lower in people who believe they have previously had the virus” is important, it is not yet entirely helpful in informing us about the type of communication that is needed in this group.

## Reaction 2
> [name=ChrisShen] I think its a very interesting connection between this research and the Health Beliefs model. The finding that most people did not perceive a great threat to themselves however still thought they had COVID was one of the more interesting findings in this paper.
>
I was surprised to see immunity of COVID-19 so low. Wouldn’t that be the driving factor behind lack of adherence to protective measures? This tells me, therefore, that there must be another reason behind a lack of adherence to protective measures, as it doesn’t make sense to me why you would think you have had COVID and still go out, without the belief of immunity or risk to others?

## Reaction 3

I wasn’t convinced by the author’s dismissal of age and factors in the results seen of lack of adherence. I wonder whether age or employment factors have more to do with less adherence to lockdown measures than whether or not they believed they have COVID-19 than is let on. The paper did touch on this, saying they had some influence, but I wonder whether there would be more emphasis of age as a factor if the paper had age as its focus.

# Shaily

## Summary
Smith and colleagues aimed to determine whether those who believed they had COVID-19 changed their self-reported behaviour.

Their sub-aims included to determine whether those who believed they had COVID-19 were less worried, more likely to believe they possessed some immunity to COVID-19, and perceived lower risk of COVID-19 to themselves.

>
> [name=Joli Doan][time=Fri 17 June] I like how you included the sub-aims as they are important results that are discussed in the paper
>
> [name=junekam17]I agree ^ that the mention of the subaims was a factor I did not consider in light of the larger hypothesis, and it's great that you mentioned them considering how well they tied in with the measures used in the survey + results at the end.

> [name=ChrisShen] I think including the sub-aims was very important considering how many different questions the researchers were looking at.

The experimenters surveyed participants online using quota sampling with questions relevant to their hypotheses.

Results revealed those who thought they had COVID-19 were less likely to adhere to social distancing measures, went out more times in the last seven days, were more likely to agree they had some immunity to COVID-19, and were less worried about COVID-19.
> [name=ChrisShen] Very concise summary of the main findings which would have the greated implications. Also the most relevant considering the aims and hypotheses were specifically looking at behaviour
>
This may have implications as the number of people believing they had COVID is likely to increase, hence influencing social norms and general adherance to protective measures.
> [name=Joli][time= Fri 17 June] Really concise and well-worded summary that includes all important details. Although it kind of repeats the point, it might be good to include more in the conclusion, specifically a summarising line for the results and ideas for future research.
>
> [name=junekam17]I really liked that you included the implications this may have on the general public. Overall it was a really clear and well-worded summary, however it would have been good if you had mentioned what next steps the authors were planning on taking or what future studies could address as a result.

## Reaction 1
I was surprised that the experimenters did not find a significant effect for those who believed they had COVID-19 to have a lower perceived risk to COVID-19 to themselves, because their behaviour indicates a sense of relaxation and reduced concern yet this does not translate to an actual reduced perceived risk.

## Reaction 2
It seems that the next step in this area of research would be to investigate the behavioural consequences of believing COVID infection with less power and hence being able to better discern the meaningful implications of smaller statistically significant effects.
> [name=ChrisShen]I think further exploring the implications of these findings would be especially important not only for this pandemic but also potential ways to limit the spread of future pandemics.
>
> [name=Joli Doan] I agree that this paper is quite preliminary and needs further investigation, however, it might be good to explain what behavioural consequences you mean since lack of adherence to lockdown measures is already explored in the paper
## Reaction 3

# Chris

## Summary
UK study conducted April 2020

Aims
- Explored whether believing you had COVID affected your self-reported behaviour
- investigated whether people knew the most common symptoms

> I really like the level of detail in your response! But I think it would be good to contextualise what is being assessed when you say the experimented 'investigated whether people knew the most common symptoms' – i.e., to say that they were assessing the likelihood of misdiagnosis. [name=shaily]

Design and method

Participants (n=6149) were asked to complete a survey

Survey questions participants were asked
- did they think they already have COVID
- Were they tested for COVID
- The extent to which they agreed or disagreed with the statement “I think I have some immunity to COVID
- Over the last 7 days, on how many days have they been outside the home for resaons including shopping, work, social visits
> []
> []
- How worried were they about COVID (Likert scale)
- what did participants think were the most common symptoms
- Asked about their gender, employment, age, education level, region

Results

24.3% thought they had COVID
- 9.4% reported having a test
- of those who were tested, 42.6% had a positive test, 57.4% were negative
- 56.7% of those who had a negative test thought they had it anyway

Younger participants (those with children, employed full/part time) and those who worked in a key sector were more likely to report thinking they had COVID

18.5% agreed or strongly agreed they had some immunity to COVID
- those who thought they had COVID were more likely to agree that they had some immunity to COVID

38.9% reported going out to the shops on two or more days
- 29.8% reported going out to shops for items other than groceries/pharmacy more than once
- 14.3% reported meeting with friends and/or family
- those who thought they had covid were less likely to adhere to social distancing measures

50.8% reported being very/extremely worried about COVID

17.7% perceived a major risk to themselves
- 40.7% perceived a major risk to people in the UK

59.1% correctly identified cough + fever as two of the most common symptoms of COVID
- those who thought they had COVID were less likely to correctly identify these symptoms

Women were more likely to think they had COVID
- older participants were less likely to think they had COVID

> I think it would be beneficial to write out your summary in full sentences and to clarify what the different numbers and findings actually mean to the reader. Especially as the experimenters make clear that the individual findings and percentages are necessarily not conducive to any hypotheses nor can be reliable for a number of reasons listed as limitations in the article, but more so it is the associations found with statistically significant effects which were more relevant. I think this could be potentially confusing for a reader. [name=shaily]
>
> [name=junekam17]This is an incredibly detailed and well-reported summary, which allowed me to understand very clearly what the experimenters did and the results they found. However, I believe it would benefit from identifying the key points and summarising them more concisely. It would also be great to see an interpretation of the results, and what implications this may have. The inclusion of numbers, while extremely informational, may also not be necessary for a reader aiming to quickly find out the main aims and outcomes of the study.

> [name=Joli Doan] Great work but a bit long. For results the actual numbers aren't really needed and also need to include a conclusion section. Main points of aim and design included are good, just a little re-working of words needed.
## Reaction 1
I was confused by the analysis methods they used for the continuous and binary scales. I understood the results that they ended up with but the entire analysis section contained too much jargon for me to understand at the moment.

## Reaction 2
I think further research in this area needs to consider targeting specific groups within the population. The paper ackowledges that their sample may not be representative of the entire population so this would represent a way to overcome this limitation.
> [name=Joli Doan] Good point. But I think I have a different rationale behind why the next area needs to target a specific population, since the paper found that mostly young people believed they had COVID-19, meaning the paper was mainly investigative of a younger population.
## Reaction 3
I want to learn more about the reasons behind some of the beliefs that people who thought they had COVID had. Why did they believe they were immune and why did so many incorrectly state the symptoms?

# June

## Summary
* Aim: to study whether those who think they’ve had COVID are less likely to comply with lockdown measures
> [name=ChrisShen]while this was something being looking at it is worth going into more detail about the other questions being looked at.

* Design/method: survey measuring…
* Perceived immunity to covid
* Self-reported adherence to lockdown measures
* Worry about/perceived risk of covid
* Knowledge of covid symptoms (fever and cough)
* Results
* 24.3% of participants thought they had had covid, but only 4% had a pos test result
* Those who believed they had covid more likely to think they had immunity, less likely to adhere to lockdown measures, less worried about covid, and less likely to know the symptoms
> I'm not sure what the purpose of stating '24.3% of participants thought they had had covid, but only 4% had a pos test result' would be as it is not really relevant to the hypotheses or aims of the experiment. [name=shaily]

* Conclusion
* This contributes to transmission – must make sure that we continue to emphasise that lockdown measures are important

> I really like how concise your summary is![name=shaily]

> [name=ChrisShen] I think this is a really concise summary of the results and the general conclusion of the paper. Especially the conclusion that transmission could have been worse because people who thought they had COVID thought they were immune.

## Reaction 1
I can see that this paper relates to real world implications of public health and safety, particularly relevant as public attitudes and compliance can affect the rate of transmission. It could also have important implications in public health policy, which could impact the lives of the population in terms of health as wella s economic and social wellbeing.

## Reaction 2
I was surprised that such a large proportion of participants believed they had already had COVID-19, despite this subset of people not knowing that the most common symptoms of COVID-19 were a cough and fever. The discrepancy between this proportion (almost a quarter!) and the proportion of participants that actually received a positive result of their Rapid Antigen Tests. This was also interesting considering their lack of knowledge. The paper may have benefited from discussing why this was the case, and why participants may have thought they had had COVID-19 if they had not experienced the symptoms. Although they mentioned age, gender, and employment as factors, they did not offer any suitable alternatives.
> [name=ChrisShen] I think not knowing the specific symptoms was a huge contributor to the high proportion of participants who believed they had COVID. They may have attributed other symptoms to COVID simply because it was such a prominent issue at the time.
>
## Reaction 3
It seems that the next step in this area of research would be to investigate the sources of misinformation and miscommunication that led to such a large proportion of participants believing they had COVID-19; what led to this? Following, it would be worth it to research ways to minimise or mitigate these sources, as well as investigating ways to effectively communicate with this subset of the population surrounding the importance of adhering with lockdown measures.

# Our assessment criteria

1. My teammate pulled their weight
- i.e. attended Q&A, actively contributed during group meetings - weekly stand-up, helped to google/troubleshoot and supported others in their learning

2. My teammate interacted with the group in a way that created an inclusive environment for us to learn from each other

3. My teammate was engaged and proactive, ensuring that adequate effort was given to communication, listening, and facilitating group work.
- listened to each other and each other's ideas
- helped/supported each other when possible

4. My teammate was able to receive feedback effectively and remain professional and amicable.

5. My teammate was adaptable, able to consider others' ideas, and implement feedback.

# Our reproducibility plan

The paper we are planning to reproduce is:

Smith LE, Mottershaw AL, Egan M, Waller J, Marteau TM, Rubin GJ (2020) The impact of believing you have had COVID-19 on self-reported behaviour: Cross-sectional survey. PLoS ONE 15(11): e0240399.

- [The impact of believing you have had COVID-19 on self-reported behaviour: Cross-sectional survey](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0240399)
- [OSF Link](https://osf.io/y7eha/)

Insert reproduciblity plan (screenshot of plots/tables) below.

[Jenny's list of things we need to do](https://internship3361.netlify.app/posts/2021-03-05-example-reproducibility-plan/)

[Table of Participant Characteristics](https://journals.plos.org/plosone/article/figure/image?size=large&id=10.1371/journal.pone.0240399.t001)

[Plot of Results](https://journals.plos.org/plosone/article/figure/image?size=large&id=10.1371/journal.pone.0240399.g001)

[Table of Results](https://journals.plos.org/plosone/article/figure/image?size=large&id=10.1371/journal.pone.0240399.t002)

[Table of Adherence to Lockdown Measures](https://journals.plos.org/plosone/article/figure/image?size=large&id=10.1371/journal.pone.0240399.t003)