General information on ATLAS Z path

See moderators-document for description: PDF, or DOC, OPloT link

For each question, show the plot indicated with the question.

Question 1

Question 1: What does it mean when we see a peak in the distribution? Do you see any peaks that you did not expect?

Moderator discussion: Explain that the low mass peaks correspond to the J/psi and upsilon resonances and that the peaks at 1 TeV and 1.5 TeV correspond to simulated Z’ and Graviton events that we hid in the data!

Question 2

Question 2: Do you see a peak corresponding to the Higgs boson? Why not? (Hopefully students are aware of statistics limitations from local institute discussions.)

Moderator discussion: Explain about statistical limitations of the measurements and that H->yy is a very rare process.

Question 3

Moderators discussion: Explain that you’re now showing the results scaled up to a much larger amount of data and that now the Higgs (brown histogram) is visible over the background (green histogram)

Question 3: Do you think we could see the peak here even if it had the same color as the background? Why does it help to collect more data?

Moderators discussion: Explain that it depends on how accurately we can measure the background processes, but in this case this is the same size as the dataset that we did discover the Higgs in, so we could identify it.

Question 4

Moderators discussion: Explain the plot and that we’re now looking for Higgs bosons decaying to two Z bosons (that subsequently decay to electrons and muons).

Question 4: Why are there many more 4-lepton events than expected? (Here you can compare to the table in the summary screen or to the expected distribution.) What can you say about the composition in terms of 4e, 2e2µ, and 4µ?

Moderator discussion: Explain that the Z boson decays to e+e- and mu+mu- at equal rates, but muons leave a very distinctive signature in the detector while electrons are easier to fake.

Question 5

Moderator discussion: Show the “ll+4l+yy” plot and zoom into the high mass region (not the “ll+4l+yy overview” plot, as it does not allow one to select the range)

Question 5: What does it tell us that the particle at 1000 GeV is not seen in the 4-lepton and diphoton distributions, while the particle at 1500 GeV is seen in all distributions?

Moderator discussion: Explain that observing how a particle decays can give us lots of information about its properties. In this case we can deduce that the particle at 1000 GeV must have spin=1 (Z’) while the particle at at 1500 GeV has spin=2 (graviton).

Bonus

A great animation for the evolution of the Higgs->gammagamma signal and for the Higgs->ZZ. These demonstrations can help with explaining to students who claim they have discoverd the Higgs, that they probably will need more statistics.

General information on ATLAS W path

Students´ tasks include:

• Explore the structure of the proton by counting the number of W+ and W- events in W candidate events: Students identify W candidate events, decay products and (if possible) their electric charge, calculate ratio R±
• Search for the Higgs in l+νl-ν + 0,1 Jets final state (1000 real data events containing l+νl-ν + 0,1 Jets were mixed with a selection of 11000 real data events from 2011 (W candidate events and background). Students identify WW candidate events and measure the opening angle Δφll between both electrically charged leptons in transverse plane.

Question 1

Moderator discussion: Show the combination/summary table and explain to the students what they are looking at.

Question 1: What was your result for R±? How did the combination with the other institutes change the total result?

Moderator discussion: Discuss briefly why this ratio is 1.5 (and not 1) and what this tells us about the structure of the proton (should have been covered in detail earlier in the day).

Question 2

Moderator discussion: Staying with the same table, point out to the students the official ATLAS measurement

Question 2: Is the result compatible with the results measured in ATLAS?

Moderator discussion: Talk about statistical compatibility of the results (N.B. the students’ measurement should agree with the ATLAS one, within uncertainties, but doesn’t always!)

Question 3

Moderator discussion: Show histogram from the second drop-down menu, explain what is being plotted (angle between two leptons).

Question 3: You have measured the angle between two leptons. Let´s have a look at that. What exactly do the black data points mean?

Question 4

Question 4: How would you interpret the blue and green areas? What do they mean?

Moderator discussion: Explain that these are the predictions we get from the Standard Model about how many events to expect. Blue is what we expect from real WW events, and green is from other processes (ttbar, Z) that can mimic this process.

Question 5

Moderators discussion: Stay on the same plot, and show the Higgs contribution (red) by checking into the appropriate checkbox and click on "submit". Explain that you’re now showing the results with the Higgs contribution included?

Question 5: Can we claim a Higgs discovery? What would be necessary to claim a discovery?

Moderators discussion: In this case, no, we can’t claim the Higgs. Explain about how we would need more data in order to say for sure that the Higgs is there.

Bonus

• Animations for Higgs -> WW can be found here.

General information on CMS measurement

An introduction to the CMS measurement can be found here. The topic of the Measurement is W and Z bosons with additional particles to be found and studied. Students will use iSpy-webgl and CIMA to find/create:

• Ratios e/mu and W+/W-
• Mass of Z from dilepton plot
• J/Psi and Y peaks in dilepton plot
• Study 4-lepton mass plot with possible Higgs, ZZ, Z-gamma.

Students will:

• Distinguish W from Z boson candidates from event displays
• Use curvature of lepton tracks to distinguish W+ from W-
• Distinguish electron from muon events
• Transfer the following data to mentors by way of CIMA: Numbers of electron and muon events, Numbers of W+ and W- candidates, Invariant masses particle candidates in 2- and 4-lepton events.
• Calculate W+/W- and e/mu from the totals found in the Results section of CIMA.

Mentors (or their assistants) at the institutes will:

• Discuss the Institute result for ratios e/nu and W+/W-, discuss significance
• Discuss dilepton mass plot (upper plot in Mass Histogram section of CIMA) with students (J/Psi, Upsilon, Z)
• Discuss 4-lepton mass plot (lower plot in Mass Histogram section of CIMA) with students (Z-gamma, H, and ZZ)
• Numbers of electron and muon events, Numbers of W+ and W- candidates, ratios, and Mass plot will be visible to moderators in CIMA.

For more detailed instructions, see:

• CMS Masterclass Documentation
• Guide to CMS masterclass measurement
• CMS masterclass website.

To share results

• Go to the CIMA admin pages. (Login/Password = Admin/MasterClass)
• Please note cautions in Moderators section of Documentation
• Choose the date under Masterclasses and then go to the Results button.

Question 1

Moderator discussion: Share Combined Mass Histograms (overall mass histograms) with the students. Explain that the top plot shows the two-lepton mass, for events with two electrons or muons in them.

Question 1: Where are the peaks in the Mass Histograms? What do they represent?

Moderator discussion: Explain that we can see the J/psi, Upsilon and Z (point with cursor to each one) and that calculating the mass of the lepton pair tells us the mass of the particle.

Question 2

Moderators discussion: Explain that bottom plot shows the 4-lepton mass, and that, analogously to the 2-lepton plot, peaks in the distribution can tell us about the particles that were produced. In this case we can (hopefully!) see two narrow peaks corresponding to the mass of Zy and Higgs masses, and a broad cluster of events which we expect to be ZZ events.

Question 2: Do you have possible Higgs events in the 4-lepton plot? Where? Can we claim discovery?

Moderators discussion: Explain about requirements for claiming a “discovery” and that in this case we can see hints of the Higgs, but we’d need more data to be sure.

Question 3

Question 3: Why does there appear to be a Z peak in the 4-lepton plot as well? Since the Z does not directly decay to 4 leptons, how do we explain this?

Question 4

Moderator discussion: Share with students: Combined Results (Overall e/mu and W+/W- results). Explain what is shown in the table. Ask students to calculate final e/mu.

Question 4: What did you expect the ratio of electron events to muon events to be (approximately)? Is your result consistent with this?

Moderators discussion: Explain that W bosons are equally likely to decay to electrons as they are do muons, so we would expect that this ratio is = 1. If the student measurements do not have this result, suggest some reasons why this could be the case (electrons harder to identify in the detector than muons - maybe some teams selected more jets as electrons, or rejected some electrons because they thought they were jets).

Question 5

Moderators discussion: Ask students to calculate final W+/W-

Question 5: What is the ratio of W+ to W- bosons? What does this ratio tell us about protons?

Moderator discussion: Explain what this ratio tells us about the structure of the proton.

Scenarios (Questions to be used in the meeting style, when groups connect)

• Why are the two tracks of each V0 curved in opposite directions?
• Why is the radius of curvature of the proton bigger than that of the pion in Λ decays?
• Why don’t you see the Λ or the K0 before their decay?
• Why does the Λ not decay to two pions, like the K0?
• Why does the invariant mass have a width and is not a delta­function?

Questions that will be used in the Zoom polls

• Baryons?
• Mesons?

* Q2: Are Lambdas

• Baryons?
• Mesons?

* Q3: Why does the invariant mass have a width and is not a delta function?

• Because some of the detectors were not working properly
• Because of the intrinsic resolution of the detectors

* What causes the background in the invariant mass distributions of kaons?

• The two pions do not originate from the same kaon but appear as coming from the same secondary vertex
• One or both pions have been misidentified
• Both of the above

General information on ALICE - Looking for strange particles

The updated document (version 2022) can be found at https://twiki.cern.ch/twiki/pub/Main/InternationalMasterclassesModeratorManual/ALICE-moderators2022.pdf

Topic of the measurement

• Identify strange particles (V0s : Ks, Λ, anti-Λ) from their decay pattern, combined with calculation of their invariant mass.

• Find number of Ks, Λ, anti-Λ for different centrality regions for lead-lead data.

• Calculate yields for Ks, Λ, anti-Λ and strangeness enhancement factors by comparing to proton-proton data.

What the students do in each institute Visual analysis Using a simplified version of the ALICE event display based on ROOT, they identify and classify strange particles (V0s : Ks, Λ, anti-Λ) from their decay pattern, combined with invariant mass calculation. Each group of 2 or 3 students looks at 15 events. At the end of this first part, the tutor merges the results of all groups and produces invariant mass plots for Ks, Λ, anti-Λ. During the videoconference, they can give the mass values and width of the peaks. Large scale analysis - Find V0s in different centrality regions in PbPb collision Students analyse large datasets, by running a programme that selects V0s, calculates the invariant mass and produces an invariant mass plot; each group of 2 or 3 students is assigned a centrality region and they have to find the number of Ks, Λ, anti-Λ in this region. To do this, they have to fit curves to the combinatorial background (2nd degree polynomial) and the peak (Gaussian) and subtract.

Calculation of particle yields and strangeness enhancement factors Each group reports the number of Ks, Λ, anti-Λ they found in the centrality class that they have analysed. The results for the whole class are entered in a spreadsheet as the following

<left></left>
The number of events in each centrality region is given in the spreadsheet.

The number of participating nucleons in the collision, Npart, which is correlated with the centrality, is also given in the table for each centrality class.

The number of particles measured is less than the number of particles produced; to find the latter we need to take into account the efficiency; efficiency values, for Ks, Λ and anti-Λ, have been estimated and are given in the table.

In the spreadsheet, there are embedded formulas to calculate:

Yield : the number of particles (of a certain type) produced per interaction = Nparticles(produced)/Nevents = Nparticles(measured)/(efficiency x Nevents)

Strangeness enhancement: the particle yield normalised by the number of participating nucleons in the collision, and properly normalised by the yield in proton-proton collisions at the same collision energy.

Ks-Yield(pp) = 0.25 /interaction Λ-Yield(pp) = 0.0615 /interaction ; the same for anti-Λ <Npart> = 2 for pp

NOTE: the above yields for Ks and Λ refer to proton-proton collisions at 2.76 TeV (same energy as for Pb-Pb collisions, 2.76 TeV per nucleon pair); they have been calculated by interpolation, between measured Ks and Λ yields at 900 GeV and 7 TeV [internal ALICE notes].

With all this information a spreadsheet like the following is produced with the information for Ks, Λ and anti-Λ.

<left></left>

Embedded in the spreadsheet is a scatter plot, showing the enhancement factors for Ks, Λ and anti- Λ versus the number of participants.

<left></left>

The results of all institutes can be accessed at the url https://teacher-alice-web-masterclass.app.cern.ch/session (until 2021 :https://drive.google.com/folderview?id=0B9FfU3MPTgvGZk8wcmFaOE9yX2s&usp=sharing)

Inside this folder, there is an example spreadsheet, results-example.xls

Each institute fills in a spreadsheet with the name results-inst_name-xxxxxx.xls ( inst_name is the name of the institute, e.g. CERN, Nantes, Heidelberg… xxxxxx is the date, e.g. 03032015)

DISCLAIMER : The results in this example – and the results produced by this analysis – are based on a small dataset selected for this measurement and on a number of assumptions and simplifications; therefore they may differ from official ALICE results.

Institutes’ report and comments

Starting from 2017 the institutes no longer give a report of the results during the videoconference. This new concept was introduced because it was considered too repetitive reporting basically the same things up to five times.

By going to the google docs folder as described in the previous section, the moderator should bring to the screen and show the spreadsheet with the institute’s results, including scatter plot. He 'she can comment. Possible comments :

The number of Ks, Lambda and antiLambda (and the calculated yield) is higher for more central collisions and decreases as we go to less central collsions. This is expected since, in the most central ones, up to ~400 nuclei in total interact, which results in thousands of particles produced per collision.

Strangeness enhancement is observed (ratio >1). Ratio=1 would mean that there is no difference between collisions of nucleons of the lead nuclei and collisions of protons.

Show ALICE results.

<left></left>

The students have only measured Λ (1 strange quark); Their measurement is in agreement with the ALICE results, within errors. (In addition, assumptions were made about the efficiency – real life analysis takes much longer – things were simplified here to complete the measurement).

The other particles shown on the plot have higher content of s-quarks: the Ξ has 2, the Ω 3. Strangeness enhancement increases with the number of strange quarks in the baryon. Note

If there is a problem with accessing google docs, the institutes can use excel spreadsheets with embedded scatter plots on their local computer. They can then show them by screen sharing.

Additional information can be found at the URL

http://aliceinfo.cern.ch/public/MasterCL/MasterClassWebpage.html

the text describing the measurement also “Instructions to the Institutes” http://aliceinfo.cern.ch/public/MasterCL/InstituteInstructions2016.pdf

Temperature Calculation

An additional comment can be made by the moderators on the calculation of temperature from particle ratios, along the lines:

Our observation of the number of produced Lambdas and Kaons can serve as a thermometer of the matter which is produced in the collision.

If you had measured Lambdas + Anti_Lambdas and K0s per event in the 0-5% centrality Pb-Pb events and then corrected for detection efficiency (pt-integrated), you would have obtained the following values (preliminary ALICE analysis results, not yet published):

NK0s = 123.9 ± 7 (which is the dN/dy per event) NΛ = 28.8 ± 3

We then put the NK0s in relation with the number of produced pions per event. The following value has been measured for the 0-5% centrality Pb-Pb event

Nπ+ = 792.1 ± 44.1

We form the ratio and obtain

NK0s / Nπ+ = 0.156 ± 0.012 NΛ / Nπ+ = 0.036 ± 0.004

From the plot below we can then roughly judge at which temperatures the particles were produced:

<left></left>

The curves are produced with a so-called thermal model. It relates the relative abundance of particles with the temperatures in the fireball. Very roughly speaking, it shows that the production of particle of mass mi is proportional to ~ exp(-mi/T). We can see this also in the plot showing that the heavier lambdas are produced less often than the lighter kaons. Thus, our observation of the number of produced Lambdas and Kaons can serve as a thermometer of the matter which is produced in the collision. Our results show that the temperature of the fireball is somewhere between 120 MeV and 180 MeV corresponding to roughly 1.74×1012 K (compared to 5.778 K on the surface of the sun -- factor 1 billion). Within 2σ the two ratios are even in agreement with a single temperature.

• ALICE published results on strangeness enhancement:
  </verbatim>

Scenarios (Questions for 5 groups)

• Visual analysis: where do the clusters come from that are not associated with a reconstructed track?
• Visual analysis: did you observe more positively or negatively charged particle tracks? Why is this so?
• Visual analysis: why do most of the produced particles have low momenta?
• What is a reconstruction efficiency and why is it important?
• What is the difference between R_CP and R_AA?
• Why is R_AA < 1 consistent with energy loss in a quark-gluon plasma?

General information on ALICE R_AA measurement

For a description see the manual of the measurement: PDF. Further information can be found in the guide for tutors and the moderators' manual.

Instructions for the LHCb exercise can be found on this link together with merging script at the bottom of the software section. The exercise was moved from local version to web in 2022 so some instructions might be outdated.

The exercise has two parts: searching for D_0 in LHCb events and measuring its reconstructed mass; measuring the lifetime in a larger sample of events using selection cuts. Students will normally discuss combine results within their institute and discuss them before the videoconference. Local organizers should send moderators their results by 15:00 on the day of videoconference (json files with invariant masses and a single lifetime value). Moderators than merge the data to prepare a gif of D_0 mass distribution and a plot with lifetime measured by institutes and the PDG value. For remote MC in 2021, moderators can shorty discuss the plots and run zoom polls (questions below) or ask studnents to write to chat their findings. Sometimes institutes ask one or few students to present/share their results.

If there is an expert from pit which connects, we recommend to keep the physics discussion short to allow for long enough cavern tour and Q&A session.

Scenarios (Questions to be used in the meeting style, when groups connect)

*Question 1: How did the D0 mass distribution change with more data?

Moderators discuss statistics and importance of collecting large datasets

*Question 2: What can cause the background in the invariant mass distributions?

Moderators show the lifetime plot

*Question 3: Did your result of lifetime agree with the world average value listed in PDG tables?

Usually students measure larger lifetime than PDG. Moderators can discuss the sources of bias, using more cuts and better selection in real analyses etc.

*Question 4: How much data would you need to be as precise as the current world average? As example take your error ~5 fs, PDG world average error ~1 fs.

Moderators can discuss statistics, future experiments etc.

*Question 5: Is it a good idea to measure D0 lifetime by more than one experiment? *Question 6: What experimental effects might cause a difference between the measured lifetime and the world average?

Example moderator discussion and questions that will be used in the Zoom polls

Here is an example of physics discussion, you can adjust it and pick which poll questions in zoom you want to use. 4 polls seem about right to keep students engaged and stick to time (do not use all 6 questions).

Moderators discussion: Introduce measurement, say now we merge data from all groups and show the combined plot of D0 invariabt mass distribution

* Q1: How did the D0 mass distribution change with more data?

• Nothing changed
• The D0 peak became clearer
• The D0 peak became less clear (more background or more fluctuations)
• We found dark matter

Moderators discussion: Discuss statistics and importance of collecting large datasets

* Q2: What can cause the background in the invariant mass distributions?

• The pion and kaon do not originate from the D0 but appear as coming from the same secondary vertex
• The pion and/or kaon have been misidentified
• Both reasons can contribute

Moderators discussion: Show the lifetime plot

* Q3: Did your result of lifetime agree with the world average value listed in PDG tables?

• Yes, it was exactly the same value
• Yes, within uncertainty
• No

Moderators discussion: Usually they measure larger lifetime than PDG. You can discuss the sources of bias, using more cuts and better selection in real analyses etc.

* Q4: How much data would you need to be as precise as the current world average? As example take yor your error ~5 fs, PDG world average error ~1 fs.

• Two times
• Five times
• Twenty-five times

Moderators discussion: You can discuss statistics, future experiments etc.

* Q5: Is it a good idea to measure D0 lifetime by more than one experiment?

• Yes, it can show if some measurement has bias
• No, our results are always correct
• It depends if you have time

(Q5 can be used if there is one institute with value very different from others)

* Q6: What experimental effects might cause a difference between the measured lifetime and the world average?

• A mis-aligned detector
• Choosing the wrong proton-proton collision point as the D0’s starting vertex
• Human error
• All of the above

NEW in 2021

NEW in 2020

The CMS Masterclass has been updated for 2020 to include 4-lepton events and separate dilepton and 4-lepton mass plots.

NEW in 2019

Everything is stable. No new elements have been introduced... besides the use of VidyoConnect instead of VidyoDesktop. Updated instructions in the relevant sections!

NEW in 2018

There is one improvement: To streamline the discussion of measurements sets of questions have been developed which should be used by the moderators. There are 5 scenarios for ATLAS and CMS measurements. Each scenario consists of a plot or table and relevant question(s) that students are supposed to answer. The scenarios can be found in the section Physics Discussion of the Measurement