Streams
Download as DOCX, PDF0 likes563 views
The project investigates the effects of acidic rain on 163 streams in the Adirondack Mountains, analyzing various water quality parameters. A positive correlation was found between water pH and hardness, with limestone substrates displaying lower acidity compared to shale. The study also examines the association between biological diversity index (BCI) and pH, employing statistical methods such as t-tests and regression analysis.
![2)
> abline(lm(HARD~PH),col=4)
> cor(PH,HARD)
[1] 0.709633
There is positive and strong correlation between Ph and hardness and it has some outliers.](https://image.slidesharecdn.com/staticalcomputing-160513123245/85/Streams-4-320.jpg)
![3)
> substrates =split(PH, SUB)
> boxplot(substrates)
4) Here on the bottom of this picture L represents “limestone”, M represents “mixed”, and S represents
“shale”.
This picture shows that Limestone streams are less acidic. Mixed streams pH levels are nearly between
7.0 and 7.5 and Shale streams are more acidic.
5)
> L = substrates["L"] //”Type = list” We can not apply this code to t-test because it doesn’t return double type.
> L = PH[SUB == "L"] //Type = double
> S = PH[SUB == "S"] //Type = double
> M = PH[SUB == "M"] //Type = double
> t.test(L,s,alt=c("greater"))
Welch Two Sample t-test
data: L and s
t = 16.303, df = 133.13, p-value < 2.2e-16
alternative hypothesis: true difference in means is greater than 0
95 percent confidence interval:
0.6607197 Inf
sample estimates:](https://image.slidesharecdn.com/staticalcomputing-160513123245/85/Streams-5-320.jpg)
![> BCI.rmna <- na.omit(BCI)
> plot(BCI.rmna,Residual,xlab = "BCI",ylab = "Residual",main="Residual Plot BCI")
> abline(0,0)
This plot is thicken because of histogram of residuals is support this.
c)
8)
> coeffs = coefficients(lm(BCI~PH))
> coeffs
(Intercept) PH
2733.3662 -197.6937
9)
> predict.ph = coeffs[1] + coeffs[2]*7.2
> predict.ph
(Intercept)
1309.972](https://image.slidesharecdn.com/staticalcomputing-160513123245/85/Streams-8-320.jpg)
Streams
- 1. Project name: STREAMS Instructor: Assoc. Prof. Dr. Kürşad Tosun Statistical Computing CENG 3004 Ramadan Şanlı 120709042 Yavuz Selim ABAZAOĞLU 120709012 MARCH 09, 2016
- 2. Project Streams? Researchers collected samples of water from 163 streams in the Adirondack Mountains to investigate effects of acidic rain. They recorded a number of biological, chemical, and physical variables, including the stream name, the substrate of the stream (the composition of soil and rock over which they flow – limestone, shale, or mixed, the pH (acidity) of the water, the temperature (° C) of the water, the BCI ( a measure of biological diversity), and hardness of the water. 1. What does a scatterplot show about relationship between the water’s pH and its hardness? 2. 2. Is it appropriate to summarize the strength of association with a correlation? Explain. 3. Make a side by side boxplot of the pH of the streams by substrate(limestone, mixed, or shale) 4. Describe what you see in the comparative boxplots. A lower pH means the water is more acidic. 5. In this question you will compare the mean pH level of streams with limestone and shale substrates by using two sample t-test. a. Do you think the assumptions for inference are met (Independent Group Assumption, Independent Assumption, and Nearly normal Condition?) Explain. b. Perform a 4-step hypothesis test (two sample t-test)? 6. Make a scatterplot of BCI against pH. Describe what you see in the scatterplot . 7. Is there any evidence of an association between BCI and pH? a. State the null and alternative hypothesis for this test. b. Make a histogram of the residuals, Normal qq plot of residuals, and a residuals plot (predicted vs residuals). Do you think the assumptions for the inference are met(straight enough condition, Independence assumption, Does the plot thicken? Condition, Nearly normal Condition)? Explain c. State a conclusion based on the regression analysis( use an appropriate 4-step test.
- 3. 8. Write the equation of the regression line to predict BCI from pH. 9.Predict the BCI if the pH is 7.2.???????????? Statistical Computing Project Streams final report 1) > plot(PH,HARD,main = "Scatterplot between Ph and Hardness",ylim = c(30,520))
- 4. 2) > abline(lm(HARD~PH),col=4) > cor(PH,HARD) [1] 0.709633 There is positive and strong correlation between Ph and hardness and it has some outliers.
- 5. 3) > substrates =split(PH, SUB) > boxplot(substrates) 4) Here on the bottom of this picture L represents “limestone”, M represents “mixed”, and S represents “shale”. This picture shows that Limestone streams are less acidic. Mixed streams pH levels are nearly between 7.0 and 7.5 and Shale streams are more acidic. 5) > L = substrates["L"] //”Type = list” We can not apply this code to t-test because it doesn’t return double type. > L = PH[SUB == "L"] //Type = double > S = PH[SUB == "S"] //Type = double > M = PH[SUB == "M"] //Type = double > t.test(L,s,alt=c("greater")) Welch Two Sample t-test data: L and s t = 16.303, df = 133.13, p-value < 2.2e-16 alternative hypothesis: true difference in means is greater than 0 95 percent confidence interval: 0.6607197 Inf sample estimates:
- 6. mean of x mean of y 7.681818 6.946377 a) This assumption is independent assumption. This streams are occurred in the same acidic rains but their substrate are different. b) 1. H0: μL – μS = 0 HA: μL – μS > 0 2. t = 16.303 3. p-value = 2.2 x 10−16 4. Because of the small P-value, we reject H0. The mean reading comprehension score for the streams with Limestone is significantly higher than the mean score for the shale. 6) > plot(BCI~PH) > abline(lm(BCI~PH),col=4) This graph show that negative correlation between Ph and hardness biological diversity and it has some outliers. 7)Kanıt var mı?
- 7. a)#Hipotez Kurulacak?# b) > Residual = resid(lm(BCI~PH)) > hist(Residual,main = "Histogram of residual(PH,BCI)") > qqnorm(Residual,main = "Normal QQ Plot of residual")
- 8. > BCI.rmna <- na.omit(BCI) > plot(BCI.rmna,Residual,xlab = "BCI",ylab = "Residual",main="Residual Plot BCI") > abline(0,0) This plot is thicken because of histogram of residuals is support this. c) 8) > coeffs = coefficients(lm(BCI~PH)) > coeffs (Intercept) PH 2733.3662 -197.6937 9) > predict.ph = coeffs[1] + coeffs[2]*7.2 > predict.ph (Intercept) 1309.972
- 9. THANK YOU!