Riverbend City ® Activity
Conducting Data AnalysisIntroductionMentor TalkCheck Your EmailConclusion
Introduction
Welcome back to your virtual internship atthe Riverbend Community Action Center!So far, you have been introduced to youroverarching project of using data analyticsto help RCAC evaluate the effectivenessof their Ruby Lake Teen HomelessnessTask Force and figured out how to storethe data using the appropriate datamodels. Now it’s time to learn how toactually analyze the data that’s beenstored.It's time for another meeting with your mentor, Brenda.
Mentor Talk
Riverbend CityCommunity ActionCenter: Mentor'sOfficeCheck in with with yourmentor, Brenda.All right, welcome back! I hope you’re finding our talks useful.Today’s session definitely should be… we’re going to betalking nuts and bolts here.
At this point, we’ve got the data stored in an organized way sothat we can find what we need. Now it’s time to do someactual data analysis using statistics. The results of thestatistical analysis are what we need to understand what’shappening, based on the data, and make somerecommendations.
Today we’re going to walk through how to do one statisticalanalysis, an Independent Samples T-Test, using MicroSoftExcel. You can do these types of analyses with programs likeSPSS and SAS, but since we already invested in buying theMS Office suite, and Excel can run these calculations too, wejust use Excel. Using Excel takes some practice, but keep at it!It’s not that complicated once you get used to the program.
You know one of the main reasons teens can end uphomeless is because they’re having a lot of problems withtheir parents. So when the teens first come to us for services,
one of the measures we give them in their intake packet is aParental Rejection Scale. The final score on the scale is anumber that can range from 0 (no sense of rejection fromparents) to 100 (total rejection by parents). This gives us asense of how much of an issue this may be for the teenager.
I’m sure you also know that there’s research out there thatsays LGBT teens are more likely to run away from home—orget kicked out of their homes—because they feel rejected bytheir parents. We want to see if that’s the case here at theRCAC or not. So we randomly selected the Parental RejectionScale scores of 30 LGBT teens and 30 heterosexual teenswho’ve come for services over the past year.
Remember, an Independent Samples T-Test compares themeans of two different groups. Here, we have twoindependent groups: one group of LGBT teens and one groupof heterosexual teens. We have the score for each group.Once we tell Excel to run the t-test, it will do all the rest for us.
You can see that there are two columns here: one labeledLGBT and one labeled Heterosexual. In each column you cansee there are lists of numbers. Those numbers are each teen’sscore on the Parental Rejection Scale. So the first LGBT teenprovided a score of 79; 79 out of 100 suggests this teen feltpretty rejected by their parent. For the first Heterosexual teen,that person reported a score of 76; 76 out of 100 suggests thatteen also felt pretty rejected by their parent.
Looking down the columns of scores, they don’t look all thatdifferent, do they? Well, this is why we do statistical analyses:we don’t know if there’s a statistically significant differenceuntil we do the analysis! We can’t just trust our firstimpressions.
Now, we’re going to start telling Excel to run an IndependentSamples T-Test. Remember that you don’t have to calculatethe means of the group yourself: Excel will do all of that. All
you have to give Excel is the raw data (in this case, the scoreson the test) in the correct columns.
Here’s something to remember: if you tried opening up yourExcel and don’t have the Data Analysis options available, thenyou need to install the Analysis ToolPak. This is available as afree add-in in Excel: I’ll give you some instructions later onhow to do that.
Now, let’s run this t-test!
Go up to the top menu and click on “Data.”
Look to the left under the Data menu and click on DataAnalysis.
You’ll see that we have two options for an IndependentSamples T-Test:
t-test assuming equal variances, andt-test assuming unequal variances.
We have to know up front if the variances in our two groups(LGBT teens and heterosexual teens) are the same ordifferent. In other words, is the range in scores (in other words,the variance) for the LGBT teens similar to the range in scoresof the heterosexual teens or not?
Luckily, Excel will calculate that for us, too!
If you scroll up through the list of data analyses, there’s anoption for “F-test Two Sample for Variances.” Select thatoption and click OK.
A new box comes up, asking you for input. Variable 1 range isgoing to be the scores for the LGBT teens. You don’t have toinput each and every value for all thirty teens! Instead, justhighlight the first value in the LGBT column (row #2), hold theShift key, and scroll down to the final value in the samecolumn (row #31).
Make sure you DO NOT highlight the first row, with the LGBTin it: that’s not a numerical value!
The column should now be highlighted. Now MS Excel inputsthose values. You should see the Variable 1 box now hasvalues in it.
Go to the Variable 2 box and input the values from theHeterosexual column, just like you did for the LGBT teens.
Then click OK.
On the Excel spreadsheet (it might go to a different sheet orput the values in the original sheet, depending on your versionof Excel), you’ll see the results of the F-Test. If the resultscome up and the columns go missing, look at the bottom ofthe Excel sheet and see if you’re on Sheet 1 (where thescores are) or on Sheet 2. If you’re on Sheet 2, your scoresare on Sheet 1.
Look at the row with the “P(F<=f).” If the value for that is lessthan .05, then we assume that variances are NOT equal. Thismeans we would run an Independent Samples T-Testassuming unequal variances.
However, our p = 0.49. This is more than p = .05, so we canassume our variances are equal.
We will run the Independent Samples T-Test assuming equalvariances.
Now that we have that settled, let’s run the t-test.
Go back to the sheet with our LGBT and Heterosexualcolumns (look at the sheets at the bottom of the screen and goback to Sheet 1).
Select Data from the menu.
Select Data Analysis.
Select Independent Samples T-Test assuming equalvariances.
Just like you did for the F-test, highlight the LGBT columnvalues for Variable 1 Range.
Do the same thing for Variable 2 with the Heterosexualcolumn.
Click OK.
OK, then! Let’s take a look at the results.
The mean tells you the average of each group. For LGBTteens, the mean parental rejection score was 83.37. Forheterosexual teens, the mean parental rejection score was79.53. This means that both groups felt rejected by theirparents, but the LGBT teens felt more rejected compared totheir heterosexual counterparts.
The t-value is 2.47. What does that mean? What you need toknow is whether or not that t-value is statistically significant(i.e., is the difference between the means of the two groupsstatistically significant?).
This is where significance testing is key. That is determined bythe p-value; this is significance testing. Look at the P (two-tailed) value. Typically values below .05 are typicallyconsidered significant (.10 and .01 are other commonsignificant levels). In this case, the p-value is 0.02, which isbelow .05, so the difference between the two groups isstatistically significant.
OK. I hope that made sense. I know it was a lot of steps toabsorb, but I promise, this gets easy the more you do it. Andit’s a very, very powerful tool to have in your toolbox.
Data for AnalysisLGBT Heterosexual
79 76
83 72
91 77
88 71
84 68
72 81
81 93
72 77
80 71
84 82
93 74
84 79
87 82
90 86
82 81
85 83
77 74
86 81
80 91
84 79
79 80
73 78
LGBT Heterosexual
95 82
86 78
91 84
82 81
78 77
84 89
92 87
79 72
Check Your Email
You have an emailfrom your CACMentor, Brenda.Subject: Content Analysis
From: Brenda Campbell, CAC Mentor
I hope the t-test discussion made sense! There’s another typeof analysis that I wanted to run by you… My meetingschedule’s too crazy for us to talk about it face to face, but Itried to put it together as a step-by-step guide.
These days, people are very concerned with statisticalanalysis, like the t-test, because the data is considered moreobjective. And data can tell us a lot about what is happeninggenerally for a large group of people, so there’s no question
it’s useful. We know from our t-test results that LGBT teensperceive significantly higher levels of parental rejection thanheterosexual teens, but we also know that both groups ofteens perceive high levels of parental rejection. That’s generalinformation that tells us this is a problem.
But when you’re designing programs, you need to know moreabout the details of people’s experiences. We can’t get thatwith statistical analyses. We can get that with contentanalyses. “Content analyses” simply refers to the analysis ofdata that is not numerical. Qualitative researchers at thedoctoral level do detailed content analyses: we’re not going todo anything as detailed as what they do.
In addition to giving measures for the teens to complete, wehad a focus group of teens where we asked them specificallyabout their experiences with their parents.
Let’s look at part of a transcript of a focus group of teen clientswho were involved in sex trafficking. The two questions wehave here asked them about how they got along with theirparents and how they ended up leaving home. (You candownload the transcript if you’d like.) Now, what most peopledo with this type of data is simply summarize it. That’s notmaking good use of this data. We want to analyze it moresystematically, which is what a content analysis is.
We’re going to do the content analysis in five basic steps:
1. First, we’re going to identify key words and phrases thatseem important.
2. Second, we’ll just make a list of the highlighted phrases.At this point, we’re moving away from summary toactually treating what was said as data.
3. The third step is to consider which phrases seem tomean the same thing: refer to the same concept.
4. The fourth step is identifying themes. “Themes” isanother word for “concept”: we say the concepts arethemes.
5. Finally, let’s analyze those themes and see what deeperinformation we can extract.
OK! Let’s walk through this: Qualitative Analysis of FocusGroup
The purpose of a content analysis is to go beyond simplysummarizing what the respondents said. The point of acontent analysis is to find deeper meaning through asystematic analysis: that deeper meaning is typicallydiscussed in terms of identifying themes. Here are the steps toa basic content analysis.
STEP ONE:Identify keywords/phrases thatseem important.Some words are highlighted.
Transcript of Focus Group of TeenClients Involved in Sex TraffickingFIRST QUESTION: How do you get along with yourparents?
Heterosexual female(age 15):
My mother and me- wenearly kill each other ifwe’re around each otherfor too long. Like, she’s al-ways gettin’ on me aboutsomethin’. ‘Do your home-work. Why you failin’ every-thing? You need to get ajob to pay for all that s—you want.’ (Shrugs) So Igot a job- on the streets.Turnin’ tricks pays a lot bet-ter than frying fries.
Transgender female (age16):
My mother don’t knowwhat to say to me. My dad-when I see him. (Shakesher head) He says I’m con-fused. My mom says I’mjust messed up. It don’tmatter what I do- I makegood grades, don’t get intotrouble, follow my mom’srules, but- as long as I likegirls, they aren’t gonna behappy. It got to the point- Icouldn’t stay there; it wasjust too tense.
Lesbian female teen (age16):
Same here: my mom don’tknow what to do with me.My dad- he’s just- he’s dis-gusted. When he foundthat skirt in my room, helost his s—, physicallypushed me down the stairsand out the door. I fell outthe door on my back‘cause I didn’t want to hithim- he’s my dad, but-hewas through, through withme.
Gay male teen (age 15):
I was hanging on with myfriend- well, my parentsthought he was just myboy, but he was more thanthat- and we were foolingaround one day afterschool- nothing serious,just kissing and stuff- butmy mom came home earlyand caught us. I’ve neverseen her like that. She washollering and her eyeswere about to bulge out ofher head. She told me toget out of her house. Mydad- I went to spend thenight at my friend’s house,but I saw my parents thenext day- he said that mylittle brothers might getconfused if they saw me,so it was better for me notto be around there, like, allthe time.
Transgender male (age17):
(Nods in gay teen’s direc-tion) My mom thought thesame thing: she said my lit-tle sister and little brotherswould get confused beingaround me with me wear-ing baggy pants and get-ting a buzzcut and stuff.She said I was almostgrown but they were still lit-tle, so she- she had to-they have to be her priority.(Eyes look teary).
Heterosexual male (age15):
I think my parents are, like-they want to be prisonguards or something,they’re just so strict. Theywant me to do chores allthe time, and then thestudying- they act like I’msupposed to be studying24/7. I’m not ever sup-posed to play games orwatch TV or text or do any-thing except chores andschoolwork. I don’t get intotrouble and my grades areokay, but they won’t get offme. I just couldn’t take itanymore.
SECOND QUESTION; How did you end up leaving home?
Heterosexual male:
I left. I was walking homefrom the bus stop and raninto this guy- this guy that Iknew, I’d seen him around,and he told me that heknew how I could makesome serious cash, easy.My parents won’t even giveme an allowance. I figured,why not?
Lesbian female teen:
(Nodded) I left too. Thetension was just too much.I knew someone who knewthis lady who helped kidsmake money, and even letus stay at her place. I didn’trealize what she’d makeme do to stay there, but Iran away on my own.
Gay male teen:
My parents told me I hadto go.
Heterosexual female:
I was up out of there. Icouldn’t take it no more.
Transgender male:
My mom said it would betoo- too confusing for mylittle sister and brothers if Iwas around, so I had tostay somewhere else.
Transgender female:
My dad kicked me out.Literally.
STEP TWO:List the highlightedphrases.Like I said above, here we’re moving away from summary toactually treating what was said as data.
We nearly kill each other.She’s always getting on me.Mother don’t know what to say to me.He says I’m confused.Mom says I’m just messed up.They aren’t gonna be happy.I couldn’t stay there.Too tense.He’s disgusted.He lost his s—.Physically pushed me.He was through…with me.She was hollering.She told me to get out.
It was better for me not to be around there.They (the younger siblings) have to be her priority.They’re just so strict.They won’t get off me.I just couldn’t take it.I left.I left.Parents told me I had to go.I was up out of there.I had to stay somewhere else.My dad kicked me out.
STEP THREE:Consider whichphrases seem to referto the same concept.For example, in terms of their parents’ reactions to them, someteens indicated they rejected their parents’ authority, whileother kids indicated their parents rejected them. Some teensdescribed running away; others described being kicked out.Making a chart with relevant phrases (you don’t need to useevery phrase as long as you know that each phrase you’veidentified fits into one of these concepts) can help to see this:
Concept: Phrases:
Wenearly killeachother.She’s al-waysgettingon me.Theyaren’tgonna behappy.
Motherdon’tknowwhat tosay tome.[Dad]says I’mcon-fused.Momsays I’mjustmessedup.
Concept: Phrases:
I couldn’tstaythere.Tootense.I left.I was upout ofthere.
My dadkickedme out.I had tostaysome-whereelse.
STEP FOUR:Identify themes.In a content analysis, the concepts are usually referred to as“themes.” We made a chart where we grouped phrases thatseemed to go together; now I’m going to go in and identify theconcepts that seem relevant. For instance, in the chart we puttogether, it looks like some teens referred to problemsbetween the parents and the teens. I’ve labeled that theme“Mutual tension between parents and teens.”
Concept: Phrases:
Mutual ten-sion betweenparents andteens.
Wenearly killeachother.She’s al-waysgettingon me.Theyaren’tgonna behappy.
Parents’ neg-ative reactiontowardsteens.
Motherdon’tknowwhat tosay tome.[Dad]says I’mcon-fused.Momsays I’mjustmessedup.
Concept: Phrases:
Teens lefthome.
I couldn’tstaythere.Tootense.I left.I was upout ofthere.
Teens wereput out.
My dadkickedme out.I had tostaysome-whereelse.
Other teens seemed to feel that it was their parents whorejected them. I’ve labeled that theme “Parents’ negativereactions towards teens.” The next major theme appeared tobe that some teens voluntarily left home while the final themerefers to teens whose parents put them out of their home. Thethemes above suggests that there were two main types oftension within these families (according to the teens): tensionbetween the parents and teens and tension that primarilycame from the parents in response to the teen. In terms ofhow they ended up leaving home, there were the teens whoran away and the teens who were put out of the house.
STEP FIVE:Analyze concepts fordeeper meaning.At this point, we’ve distilled some repeating elements from theteens’ responses to the questions. This allows us to drawlarger conclusions, and to do so with more rigor than if we’djust given the transcript a quick read. We can say, for instance,that many teens fell into a trafficking situation after leavinghome, but that twice as many left of their own accord than asleft because their parents forced them out. Similarly, we canconclude that among these teens, tension in the home is arepeating factor, with rough equivalence in the tensionbetween parents and teens being either mutual orpredominantly from the parents’ side.
Do you see how this analysis goes beyond simplysummarizing what the teens said (like a journalist might do) toanalyzing what they said to find the themes? This is whatseparates a content analysis from a summary. This providesinformation about the experiences that these teens have hadthat cannot be obtained from a standardized questionnaire.This is the type of content analysis that you need to conductfor your project.
Thanks for reading! I hope you find this useful.
Best,
Brenda
Transcript of FocusGroup of Teen ClientsInvolved in SexTraffickingFIRST QUESTION:How do you get alongwith your parents?Heterosexual female (age 15):My mother and me- we nearly kill each other if we’re aroundeach other for too long. Like, she’s always gettin’ on me aboutsomethin’. ‘Do your homework. Why you failin’ everything?You need to get a job to pay for all that s— you want.’ (Shrugs)So I got a job- on the streets. Turnin’ tricks pays a lot betterthan frying fries.
Lesbian female teen (age 16):My mother don’t know what to say to me. My dad- when I seehim. (Shakes her head) He says I’m confused. My mom saysI’m just messed up. It don’t matter what I do- I make goodgrades, don’t get into trouble, follow my mom’s rules, but- aslong as I like girls, they aren’t gonna be happy. It got to thepoint- I couldn’t stay there; it was just too tense.
Transgender female (age 16):Same here: my mom don’t know what to do with me. My dad-he’s just- he’s disgusted. When he found that skirt in my room,
he lost his s—, physically pushed me down the stairs and outthe door. I fell out the door on my back ‘cause I didn’t want tohit him- he’s my dad, but- he was through, through with me.
Gay male teen (age 15):I was hanging on with my friend- well, my parents thought hewas just my boy, but he was more than that- and we werefooling around one day after school- nothing serious, justkissing and stuff- but my mom came home early and caughtus. I’ve never seen her like that. She was hollering and hereyes were about to bulge out of her head. She told me to getout of her house. My dad- I went to spend the night at myfriend’s house, but I saw my parents the next day- he said thatmy little brothers might get confused if they saw me, so it wasbetter for me not to be around there, like, all the time.
Transgender male (age 17):(Nods in gay teen’s direction) My mom thought the samething: she said my little sister and little brothers would getconfused being around me with me wearing baggy pants andgetting a buzzcut and stuff. She said I was almost grown butthey were still little, so she- she had to- they have to be herpriority. (Eyes look teary)
Heterosexual male (age 15):I think my parents are, like- they want to be prison guards orsomething, they’re just so strict. They want me to do chores allthe time, and then the studying- they act like I’m supposed tobe studying 24/7. I’m not ever supposed to play games orwatch TV or text or do anything except chores andschoolwork. I don’t get into trouble and my grades are okay,but they won’t get off me. I just couldn’t take it anymore.
SECOND QUESTION:How did you end upleaving home?Heterosexual male:I left. I was walking home from the bus stop and ran into thisguy- this guy that I knew, I’d seen him around, and he told methat he knew how I could make some serious cash, easy. Myparents won’t even give me an allowance. I figured, why not?
Lesbian female teen:I left. The tension was just too much. I knew someone whoknew this lady who helped kids make money, and even let usstay at her place. I didn’t realize what she’d make me do tostay there, but I ran away on my own.
Gay male teen:My parents told me I had to go.
Heterosexual female:I was up out of there. I couldn’t take it no more.
Transgender male:My mom said it would be too- too confusing for my little sisterand brothers if I was around.
Transgender female:My dad kicked me out. Literally.
Download this Transcript (documents/Focus-Group-Teen-Clients-transcript.pdf)
ConclusionYou have completed the Riverbend City: Conducting DataAnalysis activity.
Licensed under a Creative Commons Attribution 3.0 License(https://creativecommons.org/licenses/by-nc-nd/3.0/)
