Name: _____________________ The Moldy Bread Statistical Worksheet Question of the lab: Does temperature and/or humidity effect fungal growth? Brand of Bread? ______________________________________________________________________________ Based on what you know, how do you think temperature will effect fungal growth? How do you think humidity will effect fungal growth? Note these are the general hypotheses (Think science fair). They will be the ones you focus on and report in your lab report. HA1:__________________________________________________________________ Lower temperatures will cause a decrease in fungal growth and higher temperatures will cause increased fungal growth. Direct relationship between ____________________________________________________________________ the two. _________________________________________________ HA2:__________________________________________________________________ High levels of humidity will cause an increase in fungal growth showing a direct relationship between the two. ____________________________________________________________________ _________________________________________________ We are using two treatments and comparing them to a control group. Write out the null hypotheses. Note: this is not to report in your lab report. H01:___________________________________________________________________________ Temperature change will have no effect on fungal growth. _____________________________________________________________________________ Increased humidity will not impact fungal growth. H02:___________________________________________________________________________ _____________________________________________________________________________ You can use the following table as an idea for how to write your data. Date: Zip locks (sample #) R1 R2 R3 Room Temperature Percent cover # of squares covered Refrigerated Zip locks (sample #) Percent cover # of squares covered F1 F2 F3 Humidity Zip locks (sample #) H1 H2 H3 1 Percent cover # of squares covered R4 R5 Average F4 F5 Average H4 H5 Average Observations:__________________________________________________________________ The refrigerator samples developed no mold across the 21 days while the room temperature ______________________________________________________________________________ samples developed a small/medium amount of mold and the humidity samples developed the ______________________________________________________________________________ most. Record all the averages you calculated before in the table below. This will be the table you use to make your graph. The results show amount of mold growth on each bread sample and compare it to the number of days that have passed. Although you are free to use any software you like to produce your graph, most students will use Excel. For the time being, draw your graph in the space below. Day Room Temp Refrigerated 0 3 6 9 12 15 18 Percent Fungi Growth(%) 21 2 Humidity Name: _____________________ Day STOP THE REST OF THE WORKSHEET SHOULD BE DONE WITH YOUR TA OR WHILE WATCHING YOUR LECTURE. Please copy the data given to you from your lab instructor into the table below. This is the information you will be using for your stats test. You will not be taking information from any other day. Note that you have a total of 5 samples. This value is known as N. So N=5. Calculate the average by dividing the sum (calculated in the table to the left) by N. Record this value in the spaces given to the left of the table. Recall that in statistics, ! refers to the value and !" refers the mean. Therefore the sample # will be put in for !. Sample # Room temperature Percent cover Refrigerated Humidity Percent cover Percent cover !̅!""# %&#'&()*+(& = 1 2 3 4 5 Σ &!""# %&#'&()*+(& 5 =_________________ Σ !!"#$%&"$'(") 5 =_________________ #̅!"#$%&"$'(") = Sum (S) #̅*+,%)%(- = Σ !*+,%)%(5 = ___________________ Now square each of those values, then determine the sum. Room temperature Sample # (Percent cover-Mean)^2 Refrigerated (Percent cover-Mean)^2 1 2 3 4 5 Sum (S) 3 Humidity (Percent cover-Mean)^2 Now calculate the variance (S2) for each treatment as indicated. Record the values in the table below. Recall that in statistics, ! refers to the value and !" refers the mean. Therefore the sample # will be put in for !. S ! "##$ &'$(')*+,)' = S ! "'6)78')*+'9 = S ! :,$797+; = ∑(.X ,--. /0.10234520 / 01,--. /0.10234520 2 6 ) 4/5 ∑(.X ,07289:023409 / 01,07289:023409 2 6 ) 4/5 ∑(.X ;5.8984< / 01;5.8984< 2 6 ) 4/5 Room Temperature Refrigerated Humidity Mean *12345627128 ) Variance + 9 12345627128 Standard Deviation (square root the Variance) √(+ 9 12345627128 ) Calculate the T-statistic for each pairwise comparison: # Refridgerated,RDDE FGEHGIJKLIG = |=̅ ,07289:023409= =̅,--. /0.10234520 | ? 6 >6 ,07289:023409 ?>,--. /010234520 # @ABCDCEF,H##$ &'$(')*+,)' = = @ |=̅ ;5.8984<= =̅,--. /0.10234520 | ? 6 >6 ;5.8984< ?>,--. /010234520 = @ The next step is to determine whether the t statistic exceeds the critical value for that test. You need two pieces of information to determine the critical value for the T test. First, we need to decide the significance level, or α (alpha). The significance level is the probability of rejecting the null hypothesis when it is true (i.e., saying there is an effect when there actually wasn’t one). By convention, in science, we use a significance level of 5% (or 0.05). In practical terms, 4 Name: _____________________ this means we are willing to accept a 5% risk of mistakenly concluding that a difference exists when there isn’t one. Second, we need to determine the degrees of freedom of our test, which is the number of independent observations (or pairs of observations) being compared: in this case it is N-1.Thus, your degree of freedom is 4. You now have enough information to determine the critical value for t in this experiment. To do this find the row and column with your degrees of freedom and alpha level respectively. For example, if you have 10 degrees of freedom and have decided to use α=20%=0.20, your critical value would be 1.372. Use this approach to determine the critical value for t in your experiment. Circle it on this table! 5 Your critical value for t is _________________________________. Now compare this critical value to the test statistics you calculated above. If your t value is > than the critical value in the chart, then your p value is below 0.05, and you can reject the null hypothesis. T Statistic for Treatments VS Room Temperature t Refrigerated, Room Temperature : t Humidity, Room Temperature : t ≥ critical value t ≥ critical value Thus, p-value <0.05 or t < critical value t < critical value Thus, p-value >0.05 Indicate for each comparison if the p-value is greater or less than 0.05. T refrigerated, room temperature: ________________________________________ T humidity, room temperature: __________________________________________ Indicate which of the two null hypotheses (if any) were rejected. H01: ________________________________________________________ H02: ________________________________________________________ When we reject the null hypothesis, what we are in fact doing is stating that we did detect a significant difference between the means for two treatments. If we fail to reject it, we state that we have not detected a significant difference between the two means (or, put another way, that the observed difference between the means was statistically significant). 6 Name: _____________________ Reporting statistical tests in BIOL 1510 § § § In the methods section of your lab report: after describing the experiment and how you collected your data, it is usually sufficient to simply state that you used a T-test to determine whether the mean percent growth differed significantly among treatments, and that you assumed a statistical significance level of 5% (i.e. α=0.05). In the results section of your lab report: you should always summarize your results in a narrative format in your results section; don’t just show graphs, and don’t report your raw data either. The first thing you should do is summarize the means (± the standard deviation) for each treatment. You should then orient the reader towards any supporting figures that you have produced. Lastly, you should summarize the results of your statistical tests. Please use the requirements from the rubric below to guide your writing: Average for each treatment (3pts), StDev. for each treatment (3pts), t-statistics (2pts), P-value, crit. t-value, deg. of freedom (3 pts), Rate of growth from trendline equation (3pts), Clearly explains results (2pts). For this lab your results paragraph should only consist of your narrative and your line graph. In the text box, write out a first draft of the narrative results of your lab report, including the statistical results: The average mold production for the room temperature samples were 28.04, while for the refrigerated condition it was 0, and for the humidity condition the average was 61.84. Additionally, for the room temperature condition the standard deviation was 5.129, and that of the refrigerated condition was 0, while the standard deviation of the humidity samples was 3.09. Since standard deviation represents the variation in the data, the small standard deviations mean that the class was precise in their data collection, but shows more precision in their collection of the refrigerator samples than the room temperature and humidity samples. The most standard deviation is present in the room temperature condition, meaning the data had the most variation in the class and was collected in the least precise manner in comparison to the other 2 conditions. Moreover, the standard deviation was used to calculate the t-test statistic for the room temperature and refrigerated pair, which was found too be 12.24 and for the room temperature and humidity pair, which was 12.50. Using the t-test statistics and the 4 degrees of freedom, the critical value was found to be 2.776. Since both t-test statistics were higher than the critical value, the p-value is less than 0.05 for both as well. This means that both the null hypotheses can be rejected and the data is statistically significant. The graph depicts the fungal growth of the refrigerated condition to be none, while in humidity the growth rate is the fastest. In room temperature there is a medium amount of fungal growth on the bread, less than in humidity but more than in the refrigerator Room temp 7 Refrigerated Humidity